Characterization of a Chromosomal Duplication

CHARACTERIZATION OF A

CHROMOSOMAL DUPLICATION IN THE PLP1 LOCUS:

A NEW MOUSE MODEL OF PELIZAEUS-MERZBACHER DISEASE

Kristi Clark

A thesis submitted to the Faculty of the University of Delaware in partial fulfillment of the requirements for the degree of Master of Science in Biological Sciences

Spring 2011

CHARACTERIZATION OF A

CHROMOSOMAL DUPLICATION IN THE PLP1 LOCUS:

A NEW MOUSE MODEL OF PELIZAEUS-MERZBACHER DISEASE

by Kristi Clark

Approved: ______Grace M. Hobson, Ph.D. Co-Professor in charge of thesis on behalf of the Advisory Committee

Approved: ______Patricia A. DeLeon, Ph.D. Co-Professor in charge of thesis on behalf of the Advisory Committee

Approved: ______Randall L. Duncan, Ph.D. Chair of the Department of Biological Sciences

Approved: ______George H. Watson, Ph.D. Dean of the College of Arts and Sciences

Approved: ______Charles G. Riordan, Ph.D. Vice Provost for Graduate and Professional Education

ACKNOWLEDGMENTS

I would like to thank my advisor, Dr. Grace Hobson, for allowing me to join her laboratory and work on this amazing project during my graduate studies. Thank you to Dr. Erica Selva and Dr. Patrica DeLeon for taking the time to be a part of my committee. I would like to thank my fellow lab members for their support: Karen Sperle for much instruction and guidance, Heather Keskeny and Linda Banser for their generous help, Lauren Sakowski for her help and continuation of this project, and Stephanie Tantzer for not just being a great help in the laboratory, but also being a faithful friend. I must also thank the Laboratory of Dr. Robert Skoff, specifically Denise

Bessert, for the time you put into helping me with this project and all your instruction. I want to thank the Bio-Imaging center at University of Delaware, specifically Dr. Kirk Czymmek and Shannon Modla, for all your support and instruction. I wish to thank my fellow SNERPs for helping me every step of the way throughout my graduate career, especially Hunter Stitik and David Wu. Without your generosity and huge hearts I would not have made it this far. Finally to my amazing friends and family, I want to thank you for all your love and support that continually motivates me to chase my dreams.

iii

TABLE OF CONTENTS

LIST OF TABLES ...... vi LIST OF FIGURES ...... viii ABSTRACT ...... ix

Chapter

1 INTRODUCTION ...... 1

1.1 Pelizaeus-Merzbacher disease; clinical and physical phenotype ...... 1 1.2 Myelination and CNS function ...... 3 1.3 Genetics and molecular basis of disease ...... 6 1.4 Animal models ...... 9

1.4.1 Existing transgenic models of PMD ...... 10 1.4.2 The Plp1dup mouse model of PMD ...... 14

2 MATERIALS AND METHODS ...... 19

2.1 Mice ...... 19 2.2 Genotyping ...... 20

2.2.1 Nucleic acid isolation and purification ...... 20 2.2.2 Genotyping analysis ...... 20

2.3 RNA expression ...... 23

2.3.1 Nucleic acid isolation and purification ...... 23 2.3.2 Semi-quantitative and real-time quantitative RT-PCR ...... 24

2.4 Protein expression ...... 27

2.4.1 Protein isolation ...... 27 2.4.2 Western analysis ...... 28

iv 2.5 Pathology ...... 29

2.5.1 Histology ...... 29 2.5.2 Confocal and electron microscopy ...... 30

2.6 X-ray analysis ...... 32

3 RESULTS ...... 34

3.1 Validation of duplication ...... 34 3.2 Plp1dup mice have a distinct physical phenotype due to a chromosomal duplication in the Plp1 locus that includes the entire Plp1 gene ...... 35 3.3 A chromosomal duplication in the Plp1 locus that includes the entire Plp1 gene alters expression of PLP1 ...... 37

3.3.1 The duplication alters Plp1 mRNA expression ...... 37 3.3.2 The duplication alters PLP1 protein expression at P12 ...... 38 3.3.3 The duplication does not alter the Plp1/(Plp1+Dm20) splice ratio ...... 39

3.4 A duplication in the Plp1 locus that includes the entire Plp1 gene alters expression of four of the five other genes found within ...... the duplication ...... 39 3.5 A duplication in the Plp1 locus that includes the entire Plp1 gene affects expression of other myelin proteins ...... 41 3.6 A duplication in the Plp1 locus that includes the entire Plp1 gene disrupts normal myelin formation ...... 42

4 DISCUSSION ...... 62

APPENDIX ...... 74

REFERENCES ...... 79

LIST OF TABLES

Table 1.1 Summary of phenotype and Plp1 dosage of existing PMD transgenic rodents and the human disease ...... 16

Table 2.1 Primers used for genotyping and RT-PCR ...... 33

Table 3.1 Validation of duplication by capillary electrophoresis of multiplex PCR, analysis of gel in figure 3.1b ...... 47

LIST OF FIGURES

Figure 1.1 Diagram of PLP1 ...... 17

Figure 1.2 Diagram summarizing targeting steps in making of the Plp1dup mouse ...... 18

Figure 3.1 Validation of duplication by semi-quantitative multiplex PCR and by amplification across hprt gene ...... 48

Figure 3.2 Weight difference in Plp1dup males compared to wild-type males at six months of age ...... 49

Figure 3.3 X-ray analysis of the Plp1dup five month old male mice ...... 50

Figure 3.4 Developmental profile of Plp1+Dm20 mRNA expression in Plp1dup and wild-type mouse brain ...... 51

Figure 3.5 Altered expression of Plp1+Dm20 mRNA and PLP1+DM20 protein in Plp1dup relative to wild-type mice ...... 52

Figure 3.6 Plp1 and Dm20 alternatively spliced transcript levels in the Plp1dup are similar to wild-type ...... 53

Figure 3.7 Altered mRNA expression of the genes within the duplicating, other than Plp1, of Plp1dup mice relative to wild-type mice ...... 54

Figure 3.8 Close to wild-type expression levels of CNPase mRNA over the developmental set with lower protein levels in adult Plp1dup mice compared to wild-type ...... 55

Figure 3.9 Wild-type levels of MAG mRNA expression with altered protein levels in the Plp1dup mice compared to wild-type mice ...... 56

Figure 3.10 Lower levels of MBP mRNA and protein expression in the Plp1dup mice compared to wild-type mice ...... 57

Figure 3.11 Discontinuous myelination pattern in the corpus collosum of Plp1dup mice ...... 58

vii Figure 3.12 Discontinuous myelination pattern in the white matter tracks of the cerebellum of Plp1dup ...... 59

Figure 3.13 Disruption of normal myelin formation in Plp1dup spinal cord ...... 60

Figure 3.14 Preliminary data indicating how specific regions of the brain will be mapped for electron microscopy using confocol microscopy ...... 61

viii

ABSTRACT

Pelizaeus-Merzbacher disease (PMD) is a rare, progressive, degenerative central nervous system disorder in which coordination, motor abilities, and intellectual function deteriorate. The disease is one of the leukodystrophies, a group of disorders that affect growth of the myelin sheath. It is caused by mutations of the proteolipid protein 1 gene (Plp1), which is located on the X chromosome and encodes the most abundant protein of myelin. About 50-75% of PMD cases are due to duplications of a region of the X chromosome that includes the entire Plp1 gene. The duplications are typically in a head-to-tail arrangement and they vary in size and gene content.

Although rodent models have been developed, these models have Plp1 gene copies that range from two to fourteen, and none contain an actual genomic rearrangement that resemble those found in PMD patients. The mouse chromosome engineering resources (MICER) were used to generate the Plp1dup mouse model by introducing a duplication into the mouse genome that is similar to human duplications. The X chromosome duplication is 270 kb in size, which is within the size range of human duplications, and in addition to Plp1, it contains five other surrounding genes that are also commonly duplicated in PMD patients. In this work, the new mouse model was characterized, and it was concluded that a duplication that includes the entire Plp1 gene leads to a complex expression pattern of Plp1 and not just a simple

ix overexpression. A duplication in the Plp1 locus alters the expression of Plp1 as well as four of the other five genes within the duplication. Mice with the genomic duplication also display altered expression in other important myelin proteins leading to a disruption in normal myelin formation. Our initial Plp1dup mice characterization studies have only begun to reveal how a duplication of the Plp1 locus results in a disease phenotype.

x Chapter 1

INTRODUCTION

1.1 Pelizaeus-Merzbacher disease; clinical and physical phenotype

Pelizaeus-Merzbacher disease (PMD) is a rare neurological disorder affecting the central nervous system (CNS) white matter. Frederick Pelizaeus described a family with severe neurological impairments as being affected by a disease that was passed on by the mother, but does not affect her directly, in 1885, long before

Mendel‟s rules of inheritance (Pelizaeus, 1885). Ludwig Merzbacher investigated the same family 25 years later and described the widespread lack of myelin staining in the cerebral white matter observed by a pathological analysis of the brain of an affected man (Merzbacher, 1910). PMD is a disease with variable clinical and neuropathological phenotypes and is now known to be one of the leukodystophies

(Hudson, Puckett, Berndt, Chan, & Gencic, 1989). The leukodystrophies are dysmyelinating diseases that cause abnormal formation, destruction, or turnover of myelin (Cheon et al., 2002).

A wide range of symptoms and severity are observed among PMD patients.

Clinical phenotypes develop early in infancy: poor head and motor control, tremors, titubation of the head and neck when infant is seated, and an involuntary eye

1 movement or jumping known as nystagmus (Reviewed by Garbern, 2007). The disease is slowly progressive, and the patients will begin to develop more symptoms, such as ataxia, tremor of the upper limbs, spasticity of the limbs, athetotic movements, and cognitive impairment (Reviewed by Garbern, 2007). Symptoms are most commonly present within the first year of life, but patients live for about six decades

(Woodward et al., 2005).

Abnormal myelin formation throughout the brain is a distinct phenotype of

PMD. A non-uniform histochemical staining of the white matter in PMD patients was observed, and areas of relatively preserved myelin staining with the lack of myelin stain in other areas gave the white matter a patchy or “tigroid” appearance

(Seitelberger, 1954). An almost complete lack of normal myelination was seen in

PMD patients by a T2-weighted MRI of the brain (Caro & Marks, 1990; Cheon et al.,

2002). Abnormalities in myelin formation observed by MRI were described as discontinuous with preserved areas creating myelin islets (van der Knaap & Valk,

1989). MRI data confirmed the hypothesis that PMD is due to arrest of myelination during early development, and slow degradation of myelin may possibly occur later

(van der Knaap & Valk, 1989).

2 1.2 Myelination and CNS function

The myelin sheath is an important membrane structure in nervous system that is around the majority of axons. Myelin has been shown to be important in human development, which is evident by the array of different neurological diseases with altered myelin, such as multiple sclerosis (Baumann & Pham-Dinh, 2001). The unique composition and segmental structure of the myelin sheath allows support for the fast nerve conduction within the thin axon fibers (Baumann & Pham-Dinh, 2001). Myelin completely envelopes axons except at the nodes of Ranvier, the small gaps of less than

1µm in which axons are directly exposed (Arroyo et al., 2001). The role of the myelin sheath is to create a region of high electrical resistance and facilitate saltatory conduction by reducing current flow across the internodal axonal membrane (Arroyo et al., 2001). Saltatory conduction is the transmission of action potentials along myelinated axons from one node of Ranvier to the next. This increases the conduction velocity of action potentials without needing to increase the diameter of an axon and allows the electrical impulses of the axon to travel faster than they would otherwise

(Huxley & Stampfli, 1949).

Oligodendrocytes are the cells that produce the myelin that surrounds axons in the CNS. The myelin is synthesized by the extension and modification of the plasma membrane of the oligodendrocyte. A single oligodendrocyte can myelinate up to 50 axonal segments (Pfeiffer, Warrington, & Bansal, 1993). The mechanisms and exact signals that regulate the complex process of myelination are not completely

3 understood. The expression of major myelin proteins depends on the contact of glial cells with axons (Scherer, Vogelbacker, & Kamholz, 1992). As the oligodendrocyte cell membrane spirals around the axon, most of the cytoplasmic contents between membrane layers extrudes to form compact myelin around the internode, which serves to create a high electrical resistance. However, at the outer edges of the internodal sheath, at the paranodal loops, and the clefts of Shmidt-Lanterman are other areas where the cytoplasmic space remains enlarged (Garbern, 2007).

Myelin contains many proteins that are important in its development. Some of these proteins are used as myelin developmental markers, since immature oligodendrocyes express 2‟, 3‟ –Cyclic Nucleotide 3‟ –Phosphodiesterase (CNPase), and mature oligodendrocytes express Proteolipid Protein 1 (PLP1), Myelin Basic

Protein (MBP), and Myelin Associated Glycoprotein (MAG) (Dubois-Dalcq, Behar,

Hudson, & Lazzarini, 1986; Monge, Kadiiski, Jacque, & Zalc, 1986). PLP1 and MBP are the most abundant proteins of CNS myelin, accounting for 50% and 30% of total protein composition, respectively (Ellis & Malcolm, 1994; Hudson et al., 1989). Both

PLP1 and MBP are thought to play important roles in intra- and extracellular myelin compaction. Normal development of the neuronal cytoskeleton in the CNS neurons requires the proper formation of the compact myelin sheath (Brady et al., 1999). MBP so far is the only protein found to be essential for formation of CNS myelin because without it the compact myelin is absent. It is termed the „executive molecule of myelin‟ (Boggs, 2006). On the other hand, when Plp1 is knocked out only subtle

4 abnormalities in the compaction of myelin was observed. Degenerating axons are the more striking pathological abnormality observed when Plp1 is knocked out (I.

Griffiths et al., 1998). The absence of PLP1 protein negatively affects the maintenance of the myelin sheath (Garbern, 2007).

PLP1 is synthesized on the rough endoplasmic reticulum (ER) and is transported through the Golgi complex where it associates with other essential parts of myelin in membrane rafts (Simons, Kramer, Thiele, Stoffel, & Trotter, 2000). Raft formation is part of the initial myelin assembly. It is followed by vesicular transport of

PLP1 into the myelin membrane (Schneider et al., 2005). Mutations that cause PLP1 protein misfolding prevent correct vesicular transport though the Golgi. Instead, misfolded PLP1 triggers the unfolded protein response that induces ER associated degradation (Gow & Lazzarini, 1996; Meusser, Hirsch, Jarosch, & Sommer, 2005).

However, overexpression of normal PLP1 does not trigger this unfolded protein response, but causes an impaired raft assembly (Simons et al., 2000; Simons et al.,

2002). Excess PLP1 along with other essential parts of myelin are found within lysosomal and endosomal vesicles. Thus, a surplus of PLP1 creates an imbalance in the myelin constituents that negatively affects late stages of myelin formation in the

Golgi network (Simons et al., 2002). Myelin in these cells has improper structure due to the incorrect ratios of myelin lipid and protein components, even though they contain normal PLP1.

5 1.3 Genetics and molecular basis of disease

The disturbance of the myelin formation observed in PMD patients led to the proposal that the disease is attributed to a defect in a myelin protein (Gencic, Abuelo,

Ambler, & Hudson, 1989). Scientists began to further investigate mutations of the

PLP1 gene as the cause of PMD after it was mapped to the human X chromosome because of the apparent X linked pattern of inheritance observed in PMD patients

(Gencic et al., 1989; Willard & Riordan, 1985). A second piece of strong evidence leading to the investigation of PLP1 mutations as the cause of PMD, was the discovery that the dysmyelinating mouse mutant jimpy, which is pathologically and genetically similar to PMD, has a mutation in the Plp1 gene resulting in aberrantly spliced Plp1 transcripts (Morello, Dautigny, Pham-Dinh, & Jolles, 1986; Nave, Bloom, & Milner,

1987). Finally, an immunocytochemistry assay showed an absence of PLP1 in the brain of an 18-year-old patient with PMD (Koeppen, Ronca, Greenfield, & Hans,

1987). Genetic defects in the PLP1 gene were discovered to be the underlying causative factor of PMD when mutations within this gene were identified in several patients with the disease (Gencic et al., 1989; Trofatter, Dlouhy, DeMyer, Conneally,

& Hodes, 1989).

The human PLP1 gene occupies a 17 kb region of genomic DNA which is located on chromosome Xq22.3 (Koeppen & Robitaille, 2002). The PLP1 gene consists of seven exons, with exon three containing an internal donor splice site that is used to generate a smaller transcript, DM20 (Hudson, 2003). The PLP1 protein has

6 one intracellular and two extracellular loops with four-transmembrane domains, depicted in figure 1.1 (Garbern, 2007; Popot, Pham Dinh, & Dautigny, 1991). There is a 35 amino acid PLP1 specific region that is missing from the intra-cellular loop of

DM20, depicted by the blue shading in figure 1.1 (I. R. Griffiths, Montague, &

Dickinson, 1995). In the CNS during early embryonic development, DM20 is the major isoform expressed prior to myelination, but it is overtaken by PLP1 postnatally during myelination and into adulthood (Fujimoto, Roots, Burton, & Agrawal, 1976;

Timsit et al., 1995). The PLP1 gene is highly conserved across mammalian species; at the amino acid level there is complete identity between mouse and human PLP1.

Therefore, mouse models can be used as faithful representations of this disease, leading to greater understanding of PMD.

Point mutations, insertions, and deletions are known mutations of the PLP1 gene that cause PMD. However, the most common cause of PMD, accounting for more than 60% of cases, is a large genomic duplication (Sistermans, de Coo, De Wijs,

& Van Oost, 1998). These duplications most commonly occur in tandem, and include a large genomic segment that includes the entire PLP1 gene and neighboring genes (K.

Inoue et al., 1996; K. Inoue et al., 1999; K. Woodward, Kendall, Vetrie, & Malcolm,

1998; K. J. Woodward et al., 2005). PLP1 duplications are presumed to cause disease as a result of an overexpression of PLP1, but this has not yet been confirmed in patients with PMD (Anderson et al., 1998; I. Griffiths et al., 1998; Karim, Barrie,

McCulloch, Montague, Edgar, Kirkham, Anderson, Nave, Griffiths, & McLaughlin,

7 2007). Variations are observed in the size of the duplicated segment among families with the smallest duplication identified at 100 kb and the largest at 5 Mb (K. J.

Woodward et al., 2005). Therefore, PLP1 might not be the only gene that is overexpressed in these patients. In fact, the X-chromosome holds a number of other genes that could also be inappropriately expressed and contribute to the disease phenotype. The minimal set of genes that was found within the smallest duplicated region of a PMD patient was PLP1, GLRA4, and TMEM31. These genes could contribute to the PMD phenotype due to increased gene dosage.

The PLP1 gene duplication may not only lead to an overexpression, but may also lead to a shift in the PLP1/(PLP1+DM20) splicing balance in the direction of the

PLP1 isoform (Regis, Grossi, Corsolini, Biancheri, & Filocamo, 2009). However, the

PLP1/(PLP1+DM20) expression profile has only been analyzed by Regis et al. and they used fibroblasts from three PMD patients with an extra gene dosage of PLP1.

Cultured fibroblasts express the PLP1 gene at very low levels (Regis et al., 2009).

Therefore, these results might not be relevant since these findings were obtained using a cell type which only expresses a small amount of PLP1 transcript and is irrelevant to the disease. This type of analysis should be performed by observing the

PLP1/(PLP1+DM20) expression profile in tissue of the central nervous system due to the fact that the PLP1 expression is predominately detected in oligodendrocytes, which are only found in the central nervous system.

8 1.4 Animal models

Transgenic rodents have been generated in which only the Plp1 gene is overexpressed, and in these models the severity of the neurological phenotype is proportional to the copy number of the transgene (Anderson et al., 1998; Anderson et al., 1999; Bauer et al., 2002; Bradl et al., 1999; Y. Inoue, Kagawa, Matsumura,

Ikenaka, & Mikoshiba, 1996; Karim, Barrie, McCulloch, Montague, Edgar, Kirkham,

Anderson, Nave, Griffiths, & McLaughlin, 2007; Karim et al., 2010; Readhead,

Schneider, Griffiths, & Nave, 1994). There is no model that resembles a common genomic rearrangement found among PMD patients. A model that faithfully recreates the common tandem genomic duplication that includes Plp1 will not only provide information about PMD specifically, but also lead to a general understanding of how such duplications would lead to a disease phenotype.

Two mouse models currently exist that contain a tandem duplication, but they model autosomal disease rather than X-linked (Li et al., 2007; Walz, Paylor, Yan, Bi,

& Lupski, 2006). These mice have been used to understand how such arrangements lead to a diseased phenotype. A mouse was used to model Down syndrome, and was created using a Cre/loxP mediated chromosomal engineering technique to generate a

22.9 Mb duplication (Li et al., 2007). The Down syndrome mouse model had elevated expression of genes within the duplicated region, except for genes that are transcriptionally inactive. Elevated expression of the genes within the duplicated region gives rise to heart and gastrointestinal defects, characteristic of patients with

9 Down syndrome. The gene or genes that is specifically responsible for the heart and gastrointestinal phenotype has yet to be determined. The second mouse was made using the same chromosomal engineering technique to generate a 3.7 Mb duplication at chromosome 17p11.2 (Walz et al., 2006). Using the model, it was concluded that even though 18 other genes were within the duplicated region, extra gene dosage of

Rai1 was responsible for the physical and behavioral phenotype abnormalities in mice.

Our newly characterized mouse model is valuable for the understanding of how genomic rearrangements lead to a disease phenotype, and also will be a valuable model for understanding the skewing of X-inactivation in females since this cannot be tested in existing models.

1.4.1 Existing transgenic models of PMD

The current rodent models that contain extra copies of Plp1 were made by random integration into the genome of a fragment of a genomic cosmid clone that contained Plp1. All of these models carry extra copies of the Plp1 gene on an autosome, so mice can be either hemizygous or homozygous for the transgene. The different rodent models of PMD are summarized in table 1.1 and are compared to the human duplication which only contains one extra gene copy of PLP1.

All of the rodent models carry different numbers of Plp1 genomic copies, and they showed varying expression levels of Plp1. The level of expression does not correlate with the number of genomic copies of Plp1 in these animals. Since there is

10 no correlation, this makes it difficult to make predictions about what expression levels might be seen in human patients using these models. For example, the 4e mouse model has two (hemizygous) or four (homozygous) copies of the wild-type Plp1 gene.

In the 4e heterozygous mouse, Plp1 mRNA levels were expressed at 132% of the wild-type expression levels at postnatal day 19 (P19) (Kagawa et al., 1994). However, the 4e homozygous mouse has 60% Plp1 mRNA expression compared to wild-type

(Kagawa et al., 1994). These results suggest that increased gene dosage of Plp1 causes an overexpression of mRNA, but a greater increase of gene dosage will begin to decrease expression level. Unfortunately, this hypothesis is rejected when these results are compared to the #66 mouse line that harbors seven (hemizygous) or fourteen

(homozygous) copies of the wild-type Plp1 gene. Based on what was observed in the

4e mouse line, a hypothesis could be made that both the #66 hemizygous and homozygous mice will have lower levels of Plp1 transcript when compared to wild- type. However, the results showed that the hemizygous mice had 122% and the homozygous mice had 60% Plp1 mRNA expression compared to wild-type at P20

(Karim, Barrie, McCulloch, Montague, Edgar, Kirkham, Anderson, Nave, Griffiths, &

McLaughlin, 2007). In sum, with two and seven copies of Plp1 there was an overexpression of transcript, and with four and fourteen copies of Plp1 there was an underexpression of transcript. These results do not show any correlation between gene dosage and RNA levels of Plp1. This could be due to the fact that results from the #66

11 mouse line were obtained from spinal cord tissue, while the 4e mouse line data were obtained from brain.

The #66 mouse line was the only animal model that was used to obtain protein expression levels of PLP1 in the overespressor animal models. A small increase was observed in PLP1 protein expression in spinal cord from P3 #66 hemizygous mice compared to wild-type (Karim, Barrie, McCulloch, Montague, Edgar, Kirkham,

Anderson, Nave, Griffiths, & McLaughlin, 2007). An increase of PLP1 protein was also observed the #66 homozygous mice compared to wild-type. However, at a later age, P20, the #66 hemizygous mice expressed normal levels of PLP1 protein, while the

#66 homozygous mice expressed lower levels compared to wild-type. These results suggest that increased gene dosage of Plp1 does not alter protein expression levels until a greater increase of gene dosage, which will cause a decrease in protein expression level. Recall in the #66 hemizygous mice, an increase of mRNA levels of

Plp1 were observed and normal protein levels. This suggests there is a translational or post-translation regulation of PLP1 that might occur in these animals. The remaining overexpressor models need to be tested to determine whether there is a correlation between gene dosage and protein expression levels.

The number of Plp1 copies does appear to correlate with phenotypic severity.

Results from experiments with the current rodent models showed that an increase in

Plp1 gene dosage leads to demyelination and hypomyelination phenotypes that worsen with higher dosage number (I. Griffiths et al., 1998). This correlation between gene

12 dosage and phenotypic severity was also found in humans bearing extra copies of

PLP1. Patients with three or more copies of the PLP1 gene have a more severe phenotype than patients with the duplication (Wolf et al., 2005).

Proteins important in the development of myelin have been used as myelination developmental markers to determine if the development of myelin is disrupted due to extra copies of Plp1. Expression of other myelin proteins was examined in the #66 mouse line. Increased gene dosage of Plp1 was shown to decreases the protein expression of MBP, MAG, and CNPase. In the #66 homozygous mice (14 copies of Plp1) MBP, CNPase, and MAG all are expressed at 50% or less of the protein expression in wild-type (Karim, Barrie, McCulloch, Montague, Edgar,

Kirkham, Anderson, Nave, Griffiths, & McLaughlin, 2007). In the #66 hemizygous mice (7 copies of Plp1) MBP expression is 50% of that in wild-type mice, while all the other myelin proteins are expressed at normal levels (Karim, Barrie, McCulloch,

Montague, Edgar, Kirkham, Anderson, Nave, Griffiths, & McLaughlin, 2007).

None of these current animal models replicate the human disease; they are all genetically different from PMD patients. Not only does the number of genomic copies of Plp1 range from two to fourteen, but the adjacent DNA that includes other genes and regulatory elements, is not included. This DNA that is excluded might be important in the regulation of Plp1 expression.

13 1.4.2 The Plp1dup mouse model of PMD

In this thesis a new model of PMD is characterized. This model is a more faithful representation of the human duplication at the Plp1 locus that causes PMD compared to previous rodent models. This model contains a 217 kb genomic duplication at the Plp1 locus and includes the entire Plp1 gene. This duplication is within the 100 kb to 5 Mb size range and exists in the tandem head to tail arrangement common among PMD patients. Not only does the duplication include only one extra copy of Plp1 like in the human duplication, but it also includes an extra copy of adjacent DNA including five other genes.

The Plp1dup mouse was generated using Mutagenic Insertion and

Chromosome Engineering Resource (MICER) technology (Adams et al., 2004) in three steps, summarized in figure 1.2. In the first targeting step, a vector harboring a neomycin (neo) selectable marker and the 5‟ end of the human hprt gene separated by a loxP site was inserted into a region proximal to Plp1 by homologous recombination in ES cells from 129/sv/EV mice. ES cells with the insertion were selected by neomycin resistance and used in the second targeting step. The second vector harboring a puromycin (puro) selectable marker and the 3‟ end of the human hprt gene separated by a second loxP site was targeted to a distal region of Plp1 by homologous recombination. Puromycin resistant ES cells were used in the third step. Cre recombinase was introduced to these cells to cause unequal sister-chromatid exchange at the loxP sites on either side of the Plp1 gene resulting in a tandem duplication (Yu

14 & Bradley, 2001). This recombination event also brings together the 3‟ and 5‟ ends of the hprt gene, so cells in which a duplication has occurred have an intact hprt gene.

ES cells that survived selection by hypoxanthine/aminopterin/thymidine (HAT) resistance were microinjected into C57Bl/6 blastocysts to create chimeras. Chimeras were mated with C57Bl/6J females (The Jackson Laboratory) to determine germline transmission, which could only yield carrier females with the duplication. Carrier females were then bred to male littermates to obtain F2 generation. F2 litters contain four different types of mice, since the duplication is located on the X chromosome, wild-type males, wild-type females, carrier females harboring the duplication on one X chromosome, and males that contain the duplication on their only X chromosome.

The males that harbor the duplication, called Plp1dup, will be characterized in this thesis as a new mouse model of PMD.

The resulting head to tail tandem duplication was 217 kb in size. This duplication includes the entire Plp1 gene, as well as five other genes at the Plp1 locus:

Tceal 3 (transcription elongation factor A, SII-like 3), Tceal1 (transcription elongation factor A, SII-like 1), BC065397 (transcript within ADP ribosylation factor family),

Morf4l2 (mortality factor 4 like 2), and Glra4 (glycine receptor, alpha 4 subunit, 2 isoforms). The mouse model was made in the Neurogenetics Research Laboratory at

Nemours by Karen Sperle using the services of the Transgenic Core Facility at the

University of Delaware and the Transgenic and Gene Targeting Facility at Thomas

Jefferson University.

Table 1.1 Summary of phenotype and Plp1 dosage of existing PMD transgenic rodents and the human disease.

Human duplication 4e mouse #72 mouse #66 mouse Lew.PLP rat

Male Hemi Homo Hemi Homo Hemi Homo Hemi Homo Extra copies of Plp1 1 2 4 3 6 7 14 4-7 8-14 Location of extra X-Chromosome autosome autosome autosome autosome autosome autosome autosome autosome copies E16-500% E16-700% P20, P60- small Plp1 RNA levels1 P20- 122% P20- 60% 200% P19-132% P19- 60% inc P3-small inc. P3-Increase PLP1 protein levels1 P20-Normal P20-Small dec. Onset 1 yr (9 days2) 2 months Birth 1 year 14 days 15 months Birth Normal 12 days Age Death 40-60 yr (9 mo- 15 mo2) 1 year 1 month 5 months 2 months Normal Hypomyelination, later hypomyelination, Type of Myelination demyelination hypomyelination demyelination hypomyelination demyelination demyelination demyelination demyelination Second week of Late: tremor Early: tremor after 1 year: Early: Late: hindlimb life: tremor Late: hindlimb Early: poor motor control Early: hindlimb followed by followed by weight loss, hindlimb tremors, followed by tremors, Phenotype Reported Later: ataxia, tremor, and tremors, ataxia, seizures, seizures, kyphosis, tremors, ataxia, and seizures, ataxia, and spasticity and seizures weakness, and weakness, and weakness, and ataxia, and seizures weakness, and seizures ataxia ataxia ataxia seizures ataxia Homo = homozygous; Hemi = hemizygous; inc = higher levels; dec = lower levels, blank = not reported 1Values compared to wt 2Human Age converted to mouse age based on the normal life span of human reported by the CDC, 80 years, and the life span of lab mouse reported by Jackson Laboratory, two years. The assumption was made using these numbers that every mouse year = 40 human years. References: (Anderson et al., 1998; Anderson et al., 1999; Bradl et al., 1999; Garbern et al., 2002; Garbern, 2007; Kagawa et al., 1994; Karim, Barrie, McCulloch, Montague, Edgar, Kirkham, Anderson, Nave, Griffiths, & McLaughlin, 2007; Koeppen et al., 1987)

Figure 1.1 Diagram of PLP1. The PLP1 protein consists of four transmembrane domains, two extracellular loops and one cytoplasmic loop. PLP1-specific region shows the region that is missing from the DM20 isoform due to the alternatively spliced RNA and these residues, 117-151, are found within the cytoplasmic loop. Figure used with permission of Dr. James Garbern, University of Rochester Medical Center, Rochester, NY.

17 Figure 1.2 Diagram summarizing targeting steps in making of the Plp1dup mouse. The Plp1dup mouse was generated using Mutagenic Insertion and Chromosome Engineering Resource (MICER) technology (Adams et al., 2004) in three steps. The first targeting step was insertion of a vector harboring an area of homology to a region proximal to Plp1, a neomycin (neo) selectable marker, the 5‟ end of the human hprt gene, and a loxP site was inserted into ES cells from 129/sv/EV mice. The ES cells were screened for neo resistance and surviving cells were used for the second targeting step. The second targeting step was insertion of a vector harboring an area of homology that allowed it to be inserted distal to Plp1, a puromycin (puro) selectable marker, the 3‟ end of the human hprt gene, and a second loxP site. The ES cells that survived a screening for puro resistance were used in the third step. The third targeting step was introduction of the Cre recombinase into these cells to cause a recombination event at the loxP sites on both sides of the Plp1 gene resulting in a tandem duplication. This recombination also brings together the 3‟ end of the hprt gene to the 5‟ end, so the cells with the duplication also have an intact hprt gene. ES cells that survived selection by hypoxanthine/aminopterin/thymidine (HAT) resistance were microinjected into C57Bl/6 blastocysts to create chimeras. Blue circle: position of centromere, thick red line: duplicated portion proximal of Plp1, thick green line: duplicated portion distal of Plp1, red triangle: loxP site, neoR : neo resistance, puroR : puro resistance, HATR: HAT resistance.

18 Chapter 2

MATERIALS AND METHODS

2.1 Mice

The mouse colony was maintained in the Life Science Center at Alfred I. duPont Hospital/Nemours, and kept to a diet of water and standard mouse chow ad libitum. The B6.129 – Dp(XTceal3-Plp1)1Tju mouse strain or Plp1dup was generated using 129/Sv/Ev ES cells that were engineered to harbor a duplication at the Plp1 locus (for details refer to section 1.4.2 in the introduction) and microinjected into

C57Bl/6 blastocysts to make the chimeras. Chimeras were mated with C57Bl/6 mice to produce the F1 generation. Identification of germline transmission was done by selecting pups from the F1 generation with the agouti coat color, and the selected females were mated with male littermates with black coat color to produce the mixed background F2 generation. Carrier F2 females were mated with mixed background wild-type males, either males from the F1 generation or B6129SF1/J (The Jackson

Laboratory), to maintain the line on a mixed background.

2.2 Genotyping

2.2.1 Nucleic acid isolation and purification

Genotyping was done using two methods, densitometry and capillary electrophoresis, on DNA obtained from tail-snips and spleen. Results obtained using the first two methods were then confirmed by identifying the formation of a complete human hprt gene generated during the formation of the duplication.

DNA was isolated from tail snips collected at weaning using DNeasy Blood and Tissue Kit (Qiagen, Valencia, CA) following the manufacturer‟s protocol for genotyping the mice. DNA was isolated from spleen tissue using Gentra Puregene

Tissue Kit (Qiagen) following the manufacturer‟s protocol when mice were euthanized for RNA and protein expression experiments. Nucleic acid concentrations were measured using Nanodrop spectrophotometer (Thermo Scientific, Hampton, NH), and a 260/280 ratio of approximately 1.80 was verified.

2.2.2 Genotyping analysis

Semi-quantitative multiplex PCR was performed on DNA from spleen and tail snips by completing 25 cycles of PCR using a Multiplex PCR Kit (Qiagen) according to manufacturer‟s instructions. All primer pairs used for genotyping are listed in Table

2.1. Gender was determined by amplifying a region of the Y-chromosome using the mSRY-F and mSRY-R primer pair. A region of Plp1 gene was amplified using the

20 mPLP-F6 and mPLP-R1 primer pair, and a region on the proximal end of the generated duplication (region x on figure 3.1.a) was amplified using 197O15-F and

197O15-R primer pair. The amplification of a region of Plp1 and region x were compared to the amplification of a region of the dystrophin gene (mDMD-28F and mDMD-28R primer pair), which is located a distance from Plp1, and was used as an internal single copy control. Regions of the neo and puro genes were amplified using the neo-F and neo-R primer pair and the MHPP-F14 MHPP-R5 primer pair, respectively. Since neo and puro were inserted into the genome during the engineering of the duplication, only the mice that have the duplication contain a copy of these genes. This was used as a secondary verification of the duplication genotype.

The multiplex PCR products produced from the amplification using the six primer pairs on the DNA isolated from tail snips were separated using 4% agarose gel electrophoresis. The gel was stained with ethidium bromide and net intensity of the products was obtained using the Gel Doc XR system and software (BioRad, Hercules,

CA). Copy number was determined to distinguish Plp1dup mice from wild-type mice by analyzing the net intensity or area under the peak data obtained. The test region was first divided by the dystrophin single copy control region. Then this ratio was normalized by dividing it by the average ratio calculated of two wild-type control mice. If the two test regions are duplicated, the expected normalized ratio of the region to the dystrophin control is 2, and if it is not duplicated, then the expected ratio is 1.

21 The multiplex PCR products produced from the amplification using the six primer pairs on the DNA isolated from spleen were first separated by 4% agarose gel electrophoresis to verify that the amplification was successful before obtaining quantitative data using capillary electrophoresis. Products were detected and analyzed by capillary electrophoresis using the 3130x1 Genetic Analyzer (Applied Biosystems,

Carlsbad, CA). This detection was possible due to the reverse primer in each pair being FAM labeled, and area under the peak was obtained using Peak Scanner

Software (Applied Biosystems) for analysis. Copy number was determined as above except that the area under the peak was used rather than net intensity.

As further confirmation of the genotype, it was determined that the human hprt gene was present in the mice using the DNA isolated from spleen. Using the Expand

High Fidelity PCR system (Roche, Basel, Switzerland) and a touchdown PCR technique, amplification of an intact hprt gene (hHPRT-F2 and hHPRT-R primer pair) can only occur in DNA from mice harboring the duplication, due to Cre-mediated recombination step that produced the intact human hprt gene (figure 1.2).

22 2.3 RNA expression 2.3.1 Nucleic acid isolation and purification

For RNA expression experiments, half brain cut along the sagittal plane was obtained from postnatal day 1, 7, 10, 12, 14, 21, 60 and 120 (P1, P7, P10, P12, P14,

P21, P60 and P120) male mice. This age set allowed for visualization of a developmental time course when myelin rapidly accumulates from postnatal day 10-30 and reaches a higher synthesis rate at approximately day 21 to 23 and into adulthood

(Costantino-Ceccarini & Morell, 1972). For these expression experiments, biological replicates of three wild-type and three Plp1dup mice were used. Brain tissue was isolated, snap frozen, and stored at -80 °C.

RNA was isolated using RNeasy Midi Kit (Qiagen) following the manufacturer‟s protocol. RNA was then DNase treated using TURBO DNA-free kit

(Ambion, Austin, TX) following the manufacturer‟s protocol. A Nanodrop spectrophotometer (Thermo Scientific) was used to determine nucleic acid concentration, and to ensure that 260/280 ratios were approximately 2.00. The total

RNA quality was obtained by RIN algorithm (RNA Integrity Number) using the 2100

Bioanalyzer and the RNA 6000 Nano Chip kit (Agilent, Santa Clara, CA), following manufacturer‟s protocol. If a RIN number of seven or above was obtained, then the quality of the RNA was sufficient for qRT-PCR.

23 2.3.2 Semi-quantitative and real-time quantitative RT-PCR

One microgram of total RNA was denatured of secondary structure at 65 °C for

5 minutes. The total RNA was then reverse transcribed in a 20 µl reaction at 37 °C for

60 minutes using the Omniscript reverse transcription kit (Qiagen) following the manufacturer‟s protocols. All cDNA was stored at -20°C and used for both semi- quantitative and real-time quantitative RT-PCR experiments.

Semi-quantitative multiplex PCR was performed with 1 µg of cDNA prepared from mouse brain RNA using the Multiplex PCR Kit (Qiagen). The RT-PCR was carried out using the following protocol: 15 minutes at 95 ºC, followed by 24 cycles

(P1 using 30 cycles) of 45 seconds at 94 ºC, 1.5 minutes at 60 ºC, and 1.5 minutes at

72 ºC, then ending with 10 minutes at 72 ºC. PCR products were stored indefinitely at

4 ºC. The primer pair mPLP2-F1 and mPLP4-R2 was used to amplify both Plp1 and

Dm20 transcripts, as this primer pair yields different sized products from these splice varients. To ensure that equal amounts of RNA were used, the endogenous control β- actin was amplified with the mACTb-F2 and mACTb-R primer pair. Products were detected and analyzed by capillary electrophoresis using the 3130x1 Genetic Analyzer

(Applied Biosystems). This detection was possible due to the reverse primer in each pair being FAM labeled, and Peak Scanner Software (Applied Biosystems) was used for the analyses. The splice ratio was determined by analyzing the peak areas using the following equation: (Plp1/(Plp1+Dm20)).

24 Real-time quantitative RT-PCR was performed for quantification of mRNA expression of Plp1, mRNA expression from the five other genes within the duplication, and the mRNA expression of myelin proteins Mag, Mbp, and CNPase using QuantiTect SYBR Green PCR kit (Qiagen) following the manufacturer‟s instructions. In each experiment, 0.5 µg cDNA was used for each reaction, performed in 20 µl triplicates for each sample, using three biological replicates of wild-type and

Plp1dup mice at each age. The following QuantiTect Primer Assays (Qiagen) were used according to the manufacturer‟s instructions: Mm_Plp1_1_SG,

Mm_Glra4_1_SG, Mm_Tceal1_3_SG, Mm_Tceal3_1_SG, Mm_Morf4l2_1_SG,

Mm_BC065397_2_SG, Mm_Mbp_1_SG, Mm_Mag_1_SG, and Mm_Cnp_1_SG.

The primer assays mus musculus calnexin (CANX) and mus musculus cytochrome c-1

(CYC1) were used to amplify endogenous controls (PrimerDesign, Southhampton,

UK) (see appendix for selection process). qRT-PCR was carried out using the ABI prism 9700 (Applied Biosystems, Foster City, CA) with an initial denaturation step at

95 °C for 15 min to insure complete activation of the hot start DNA polymerase, followed by 40 cycles of 94°C for 1 min, 55°C for 30 seconds, and 72°C for 30 seconds. The specificity of amplified product was confirmed by a melting curve analysis, which included an initial denaturation step at 95°C for 15 seconds, an annealing step at 60°C for 15 seconds, and finally a gradual increase to 95°C (2% ramp rate) to denature again.

25 Calculations of qRT-PCR were performed by first obtaining Ct values detected during amplification using 9700 SDS RQ software (Applied Biosystems) for each transcript. All transcrips of interest were normalized by dividing raw Ct values for each sample by the appropriate normalization factor calculated from the geometric mean of the two most stable house-keeping genes CANX and CYC1. As shown in the

Appendix, the most stable house-keeping genes were selected using geNorm application for Excel (Microsoft) which is available for public use at http://medgen.ugent.be/~jvdesomp/genorm/. Relative quantification (RQ) of RNA expression in the Plp1dup mouse was determined using the average of the three wild- type males as the calibrator DataAssist Software v2.0 (Applied Biosystems) with the 2-

∆∆CT method. The relative amounts of mRNA for each transcript were expressed as a percentage of wild-type using the average RQ values for Plp1dup mice at each time point. A program was developed using the Statistical Analysis System SAS (SAS

Institute Inc, Cary, North Carolina) to analyze the raw Ct values. Statistical significance between wild-type and Plp1dup mice RNA expression levels was determined using a standard ANOVA test. Standard deviations were calculated and used to show variation between mice.

To determine the developmental expression pattern of Plp1 in the brain, the same qRT-PCR procedure was used. However, only CYC1 was used as the endogenous control due to space constraints on a single plate. Relative quantification

(RQ) of the RNA levels at each time point was determined using the average of the 3

26 wild-type P1 males as the calibrator using DataAssist Software (Applied Biosystems) v2.0 with the 2-∆∆CT method. The relative amounts of Plp1 mRNA were expressed as a fold change relative to P1 wild-type expression using the average RQ values for each time point. This allowed for a visualization of the expression of Plp1 after P1 in wild- type and Plp1dup mice. Standard deviations were calculated as described above.

2.4 Protein expression 2.4.1 Protein isolation

For protein expression experiments, half brain cut along the sagittal plane was obtained from P1, P7, P10, P12, P14, P21, P60 and P120 male mice, as described for the RNA expression analysis. Biological replicates of three wild-type and Plp1dup mice were also used. Brain tissue was isolated, snap frozen, and stored at -80 °C.

Protein extracts were prepared from mouse brain by homogenization in suspension buffer which contained 0.1M NaCl, 0.01M Tris-Cl (pH7.6), 0.001M

EDTA( pH 8.0) and addition of 100µl of protease inhibitor (Thermo Scientific) per

10mL of buffer at time of use. An equal amount of 2x SDS Gel Loading Buffer which contained 100mM Tris-Cl (pH 6.8), 4% SDS, 20% glycerol, and 200mM DTT was added to the homogenate and mixed well by inversion and vortexing. The homogenate was sonicated ten pulses of 1 sec each, and then centrifuged at 5,000 x g for 5 minutes and supernatant liquid was removed and used as the protein lysate. Protein

27 concentration was determined by measuring absorbance at 280 using a Nanodrop spectrophotometer (Thermo Scientific).

2.4.2 Western analysis

30 µg of protein was separated on 4-20% GTX gel (BioRad) and transferred to a PVDF membrane (Invitrogen, Carlsbad, California). Membranes were probed with mouse monoclonal antibodies anti-PLP1 at a dilution of 1:2000 (AA3 Hybridoma, a kind gift from Dr. Alexander Gow, Wayne State University), anti-MBP at 1:1000

(Millipore, Cat#09-849), anti-CNPase at 1:5000 (Covance, SMI-91R), anti-MAG at

1:1000 (Millipore, Cat#MAB1567), and anti-GAPDH at 1:20000 (Sigma,

Cat#G8795). Membranes were then incubated in DyLight-405 (Jackson

ImmunoResearch), Cy3-, or Cy5-labeled (GE Healthcare, Piscataway, NJ) secondary antibody diluted 1:2500. Fluorescently labeled protein was detected using the

Typhoon Trio scanner (GE Healthcare). When DyLight-405 or Cy5 secondary was used, the 633nm wavelength laser was selected to visualize the protein, but when Cy3 secondary was used, the 532nm wavelength laser was selected. Bands were quantified using ImageQuant TL v2003.03 software (GE Healthcare) with GAPDH as a control for loading accuracy. Relative amounts of protein from Plp1dup mice were expressed as a percentage of wild-type.

28 2.5 Pathology

For pathology experiments all mice were anesthetized then perfused with fixative before brain tissue was harvest. The mice that were used in these experiments were one, three, and six months of age, and tail-snip DNA was used to identify mice needed for experiments. Biological replicates of 2-4 wild-type and Plp1dup mice were used for observation in these experiments.

2.5.1 Histology

Animals were anesthetized with 4 µl of 50mg/µl ketamine, 20mg/µl xylazine per gram of body weight by intraperitoneal (IP) injection. Each animal was completely perfused through the left ventricle with 2.5ml of 4% paraformaldehyde in

0.1M PBS per gram of body weight slowly using a pump. The brain was then isolated and placed in 4% paraformaldehyde in 0.1M PBS buffer at 4 °C. The brain was sliced with a razor blade in 2 mm coronal sections and submitted to the Histotechnology

Core at A. I. duPont Hospital/Nemours Children‟s Clinic for processing, sectioning, and staining. The sample was taken through a series of graded alcohols, cleared and infiltrated with paraffin on the Leica ASP300 tissue processor (Leica Microsystems

Inc, Buffalo Grove, IL). Tissues were embedded in paraffin on the Leica EG1160 embedder (Leica Microsystems Inc). Samples were cut in 5 µm sections on the Leciva

29 RM2255 microtome (Leica Microsystems Inc). Sections were stained with Luxol Fast

Blue-Periodic Acid Schiff‟s-Hematoxylin (LFB/PAS) procedure (Sheehan and

Hrapchak, 1980), which was used to visualize the myelin content.

2.5.2 Confocal and electron microscopy (EM)

Brian and spinal cord tissues were used to visualize myelin formation by confocal and electron microscopy to obtain a detailed view of myelination in the

Plp1dup mice compared to wild-type mice. Animals were anesthetized, as described in 2.5.1. Each animal was completely perfused through the left ventricle with 2.5ml of

4% paraformaldehyde and 0.5% glutaraldehyde in 0.1M PBS per gram of body weight slowly using a pump. After perfusion, brain and spinal cord were removed and placed in fixative for at least 24 hours.

Spinal cord was sent to Dr. Robert Skoff (Wayne State University) for EM processing and analysis. The spinal cords were dissected transversely into 1mm pieces and processed as follows. Tissues were rinsed in 0.1M phosphate buffer 3 x 10 min, fixed with 1% osmium tetroxide (Electron Microscopy Sciences, Hatfield, PA) for 2 hrs. They were rinsed again with 0.1M phosphate buffer 3 x 10 min and dehydrated with increasing percentages of ethanol for 2 x 5 min up to 100% for 3 x 10 min and then with propylene oxide (Electron Microscopy Sciences) 3 x 10min.

30 Tissues were then infiltrated overnight with a 1:1 solution of propylene oxide and

Araldite 502 (Electron Microscopy Sciences), then embedded the next day with 100%

Araldite and cured at 60 °C for 48 hrs. Ultra-thin sections were cut on an Ultracut E microtome (Reichert Technologies, Depew, NY) using a glass knife and diamond knife respectively. These ultra-thin sections were stained with uranyl acetate (Electron

Microscopy Sciences) and lead citrate (Polysciences, Warrington, PA) and visualized and imaged using Jeol 1010 electron microscope to visualize.

Brain tissues were infiltrated with 2.3M sucrose in 0.1 M Sorensen‟s phosphate buffer. Samples were then embedded in cryomolds with Tissue Freezing Medium

(Electron Microscopy Sciences), frozen at -30 °C, and stored at -80°C until sectioned.

A CryoStat (Leica Microsystems, Wetzlar, Germany) was used to prepare sections

20um thick using the CryoJane Tape-Transfer system (Leica Microsytems). Sections were stained using a BrainStain Imaging Kit (Invitrogen), which contained

FluoroMyelin green fluorescent myelin stain, NeuroTrace 530/615 red fluorescent

Nissl stain, and DMSO DAPI dihydrochloride, following the manufacturer‟s protocol.

31 2.6 X-Ray analysis

Mice were anesthetized with 3-5% isoflurane and placed in the In-Vivo

Imaging Systems (Carestream, Rochester, NY) cabinet. They were maintained with

2% isoflurane through-out the analysis period. X-ray images were obtained using the

KODAK 1D Imaging System software (Carestream Health, Inc.). Afterwards, the mice were removed from the isoflurane and monitored until completely recovered.

Table 2.1. Primers used for genotyping and RT-PCR.

Genotypting Primer Name Sequence mSRY-F 5’-GCATTTATGGTGTGGTCCCGTG-3’ mSRY-R 5’-GCTCTACTCCAGTCTTGCCTGTATG-3’ mPLP-F6 5’-CCACCTGTTTATTGCTGCGTTTG-3’ mPLP-R1 5’-GCACACTGCATTGGCACTACATTAAC-3’ 197O15-F 5’-AATGTCTGGGTAGGGTGCTTGTC-3’ 197O15-R 5’-AAGAACGGAGATGCTGTTAGATGAG-3’ mDMD-28F 5’-GCAGACAGAAAATCCCAAAGAGC-3’ mDMD-28R 5’-TTGTGCTAACTTCATTCCACCAAG-3' neo-F 5’-GATGATCTGGACGAAGAGCATCAG-3’ neo-R 5’-TAAAGCACGAGGAAGCGGTCAG-3’ MHPP-F14 5’-GCCTTCCATCTGTTGCTGCG-3’ MHPP-R5 5’-GCTGCAAGAACTCTTCCTCACG-3’ hHPRT-F2 5’-GCTGAGGATTTGGAAAGGGTG-3’ hHPRT-R 5’-CCTTGAGCACACAGAGGGCTAC’3’ RT-PCR Primer Name Sequence mPLP2-F1 5’- CTACTGGATGTGCCTGACTGTTTCC-3’ mPLP4-R2 5’- GGGAAGAAACCTCTCTGGTTTCAG-3’ mACTb-F2/R 5’- GTGACGTTGACATCCGTAAAGACC-3’ mACTb-R 5’- AAGAAAGGGTGTAAAACGCAGCTC-3’

33 Chapter 3

RESULTS

3.1 Validation of duplication

Figure 3.1 and table 3.1 show the results of two methods that were used to confirm the duplication in Plp1dup mice. In one method, semi-quantitative multiplex

PCR was performed on the spleen DNA to determine gender using a primer pair that amplifies a region of the Y-chromosome. To determine genotype, a region of the Plp1 gene as well as a region on the proximal end of the generated duplication (region x) were amplified for comparison to the amplification of a region of the dystrophin gene.

In addition, regions of the neo and puro genes were amplified to verify the duplication since they were inserted into the genome during the engineering of the duplication.

Figure 3.1a depicts where the primer pairs anneal. PCR products were separated by gel electrophoresis and stained with ethidium bromide for visualization before submitting the samples for capillary electrophoresis (figure 3.1b). Males containing the duplication were easily identified in figure 3.1b based on the amplification of neo and puro (samples in lanes 2, 4 and 5). Results showed that the mice had a greater amplification of the Plp1 gene and region x compared to the amplification of the dystrophin gene as observed by visualization of the bands. Quantitative data for these

34 PCR products was obtained by capillary electrophoresis and analyzed by measuring the area under the peak (table 3.1). The values obtained for the Plp1 region and region x were divided by the value obtained for dystrophin and then the ratio was normalized to the ratio of two wild-type control mice. If the test regions are duplicated, the expected normalized ratio to the dystrophin control is 2. If it is not duplicated, then the expected ratio is 1. Table 3.1 shows this analysis and these data indicate that pups

2, 4, and 5 must harbor the duplication. The genotyping that was performed using

DNA isolated from tail snips was similar with the exception that quantitation was performed using densitometry of the ethidium bromide-stained gels instead of by capillary electrophoresis.

For a second analysis done by gel electrophoresis, the presence of the duplication was confirmed by amplification of hprt gene using the primers, hHPRT-F2 and hHPRT-R. The hprt gene was generated during the recombination event that produced the duplication, as depicted in figure 3.1a. Figure 3.1c shows representative data of the amplified hprt product on an agarose gel. Once this product was confirmed, the mouse was identified as a Plp1dup mouse.

3.2 Plp1dup mice have a distinct physical phenotype due to a chromosomal duplication in the Plp1 locus that includes the entire Plp1 gene

At early ages, Plp1dup mice do not have a overtly abnormal phenotype, but slight differences were noticed during handling and weaning of three week old mice.

For example, while a wild-type male was being held by the tail, he would lift himself

35 up to clasp the holder‟s hand. However, the Plp1dup males have a more difficult time with this task, and some cannot even lift themselves up completely.

At six months of age, obvious weight differences were observed in Plp1dup males compared to wild-type littermates. Figure 3.2 shows that Plp1dup males at six months of age have a lower body weight compared to wild-type littermates. Plp1dup males weigh an average 35.25 g while wild-type males were slightly heavier averaging

41.22g. There was a statistically significant difference observed in the weights of the two groups of mice.

At five months of age and older, a slight curvature of the spine was observed as a hump in Plp1dup males that was not present in wild-type males. An X-ray analysis was performed on three five month old male mice, one wild-type and two Plp1dup, and the results are presented in figure 3.3. As shown by the red arrow, the hump was seen by X-ray analysis. However, in these data, two Plp1dup males were analyzed, and only one has this altered curvature in the spine. It was not clear if the other

Plp1dup male will develop this phenotype at a later age, or if the hump will never develop.

When Plp1dup males were over one year old, they develop seizures more often than wild-type littermates and usually die at this time. This appeared in five of seven

Plp1dup mice that were kept to a year old. Only two out of seven wild-type litter mates developed seizures. Therefore, this appears to be a phenotype that is more

36 common in Plp1dup males compared to wild-type. The lab is currently working on quantifying and statistically analyzing this phenotype in a greater sample size.

3.3 A chromosomal duplication in the Plp1 locus that includes the entire Plp1 gene alters expression of PLP1

3.3.1 The duplication alters Plp1 mRNA expression

The developmental expression of Plp1 in the mouse brain was investigated to determine the normal expression pattern of Plp1 during the formation of myelin and into adulthood to elucidate the effect of a duplication on Plp1 expression. Real-time quantitative PCR (qRT-PCR) was performed to observe the developmental expression of Plp1. The results of the analysis of Plp1+Dm20 at each time point relative to P1 wild-type expression is presented in figure 3.4. A triphasic pattern was observed in the wild-type mice. The initial phase consisted of a period of regular increase in the level of Plp1+Dm20 mRNA from P1 to P12. During the second phase, a more dramatic increase in total Plp1+Dm20 transcript was observed from P12 to P21. This was followed by a sharp decline in the level of Plp1+Dm20 mRNA reaching a constant level at P60. The peak of Plp1+Dm20 mRNA expression occurred at P21 in wild-type mice. However, a very different expression profile was observed in the Plp1dup mouse. An initial phase of a regular increase in the level of mRNA was observed from

P1 to P10 that does not appear very different from the wt. However, at P12 to P14, a more dramatic increase in expression of Plp1+Dm20 mRNA was observed in the

Plp1dup mice than the wild-type mice. At P21, the Plp1dup mouse mRNA was expressed at the same level as the wt. Finally, a slight decline, not as sharp as in the

37 wt, was observed at P60. Expression does not level out as in wild-type, but shows a dramatic increase in mRNA levels again into P120. These results show very different developmental expression profile of Plp1+Dm20 mRNA in wild-type and the Plp1dup mice.

qRT-PCR was performed on the developmental set of brain RNAs to demonstrate whether the total amount of Plp1+Dm20 message in the Plp1dup mouse was elevated at specific developmental time points compared to wild-type.

Quantitative data in figure 3.5a shows that Plp1+Dm20 mRNA levels were elevated in

Plp1dup to approximately 200% of wild-type expression levels. This increase was observed at P12, P14, P60, and P120. At the earlier developmental time points, P1,

P7, and P10, Plp1+Dm20 message levels in Plp1dup mice were not significantly different from wild-type. At P21, where the Plp1+Dm20 transcript peaks in wild-type

(figure 3.4), the Plp1dup mouse shows similar levels of the Plp1+Dm20 message to the wild-type mice.

3.3.2 The duplication alters PLP1 and DM20 protein expression at P12

Protein expression levels of PLP1 in the Plp1dup brain were examined by

Western blot analysis using brain homogenates with a monoclonal antibody (AA3) that recognizes both PLP1 and DM20 isoforms. A monoclonal antibody that recognizes GAPDH was used to verify loading accuracy. Results of the analysis are shown in figure 3.5b. Protein levels were too low to detect until P12 in the

38 developmental set in Plp1dup and wild-type mice. Total PLP1+DM20 protein levels were elevated at P12 in the Plp1dup mice, while levels were the same as wild-type mice at P14, P21, P60, and P120. The duplication alters the PLP1 protein expression at P12 only. This suggests that Plp1 gene expression is a controlled and regulated process.

3.3.3 The duplication does not alter the Plp1/(Plp1+Dm20) splice ratio

To determine whether the duplication in the Plp1 locus impairs the Plp1/Dm20 splicing ratio, the Plp1 and Dm20 transcripts were examined by RT-PCR in RNA prepared from brain of Plp1dup and wild-type postnatal brain. The expression level of just the Plp1 isoform divided by total Plp1 (Plp1+Dm20) was calculated and similar expression levels of the Plp1 isoform were observed in the Plp1dup mouse compared to wild-type (figure 3.6). At P1, Plp1 isoform was approximately 30% of the total

Plp/Dm20 transcript expressed in wild-type and Plp1dup mice. At P7 and older, the

Plp1 isoform was approximately 70% of the total Plp/Dm20 message transcribed in the wild-type and Plp1dup mice. The duplication does not alter the splice ratio of the

Plp1 transcripts.

3.4 A duplication in the Plp1 locus that includes the entire Plp1 gene alters expression of four of the five other genes found within the duplication

The most common cause of PMD is duplication of a region of the X chromosome that includes the entire Plp1 gene and the duplicated regions vary in size

39 and gene content; therefore, it was important to determine if the duplication affects the expression level of other genes within the duplication. The five other genes within the

Plp1dup mouse duplication were measured at a transcriptional level using qRT-PCR.

Transcription elongation factor A (SII-like) 3 (Tceal3), transcription elongation factor

A (SII-like) 1 (Tceal1), mortality factor 4-like 2 (Morf4l2), BC065397, and glycine receptor subunit alpha 4 (Glra4) message levels were obtained using RNA isolated from brains of the developmental set of Plp1dup and wild-type mice by qRT-PCR.

Figure 3.7a shows the duplication in the Plp1dup mice to help visualize the location of each gene relative the Plp1 gene.

Elevated transcript levels of Tceal3, Tceal1, Morf4l2, and BC065397 were observed in the Plp1dup mice compared to wild-type mice at all postnatal time points except P1 (figure 3.7b, c, d, and e). Expression of Tceal3 was observed to be 200-

250% of wild-type in the Plp1dup mouse at most of the developmental set (figure

3.7b). Some variation among the biological replicates lead to no significance difference in Tceal3 mRNA expression at P10 in Plp1dup mouse compared to wild- type. Levels of 200% that of wild-type mRNA was observed among the developmental set of Tceal1 and BC065397 in Plp1dup mice (figure 3.7c and e).

Levels that were 200% of wild-type was observed for Morf4l2 mRNA. However, variation among replicates lead to no significant difference in Morf4l2 mRNA levels in Plp1dup mouse compared to wild-type at P7. Interestingly, at P120 Morf4l2 mRNA levels were elevated to 350% of wild-type expression in Plp1dup mice. Unlike the

40 other genes within the duplication, a change in the Glra4 message levels was not observed in the Plp1dup mouse (figure 3.7f). A duplication in the Plp1 locus alters the expression of four of the five genes found within the duplication not including

Plp1.

3.5 A duplication in the Plp1 locus that includes the entire Plp1 gene affects expression of other myelin proteins

To determine whether changes in Plp1 gene dosage due to a chromosomal duplication affects the expression of other representative myelin proteins, CNPase,

MAG, and MBP transcript and protein levels were measured in the Plp1dup mice at the developmental time set and compared to those of wild-type. qRT-PCR was performed with the developmental set of RNA from brain and Western blot analysis with brain homogenates from Plp1dup mice and compared to wild-type mice.

The data from real time qRT-PCR shows that the Plp1dup mice have expression of CNPase message that was similar to that of the wild-type mice (Figure

3.8a). The only significant difference between Plp1dup mice compared to wild-type mice was observed at P21, where transcript levels were decreased. Figure 3.8b shows

CNPase protein expression over the developmental time course. It was expressed at

P7 and later. Plp1dup mice express statistically significant lower levels of CNPase at

P60 and P120 compared to wild-type mice.

The data from real time qRT-PCR shows that Plp1dup mice have similar MAG expression compared to wild-type mice throughout the developmental time course

41 (Figure 3.9a). MAG protein expression was found to be altered along the developmental set (figure 3.9b). Plp1dup mice express significantly lower levels of

MAG at P7, P10, and P21.

The real time qRT-PCR shows that Plp1dup mice show a slight decrease in expression of MBP mRNA compared to wild-type mice (Figure 3.10a). P7, P21, and

P60 showed significantly lower Mbp mRNA levels in Plp1dup mice compared to wild-type. MBP protein expression was not observed until P21 (figure 3.10b).

Plp1dup mice express lower levels of MBP at all the time points compared to wild- type mice, and levels of expression were significantly lower at P60 and P120. A duplication in the Plp1 locus alters the expression of MBP, MAG, and CNPase.

3.6. A duplication in the Plp1 locus that includes the entire Plp1 gene disrupts normal myelin formation

Standard histological analysis was performed to assess the extent and onset of myelin in the Plp1dup mice. Whole brains from three wild-type and three Plp1dup males at one, three, and six months were coronally sectioned and stained with Luxol fast blue and Periodic acid shift (LFB/PAS) to provide an estimation of myelin content. The LFB stain helped visualize by staining myelin blue and PAS was used as a pink counter stain.

Disruption of normal myelin formation was observed in the corpus callosum of the Plp1dup mice. Figure 3.11 shows this histological analysis, and an even myelin pattern was observed in the sections from the wild-type mice. However, a disrupted

42 discontinuous pattern was seen in the sections from the Plp1dup mice. This discontinuous pattern was observed as early as one month (figure 3.11a). At three months, this discontinuous pattern was much more dramatic in the Plp1dup sections

(figure 3.11b). Sections from wild-type mice at three months show the corpus callosum with a solid even myelin pattern as shown by the blue stain, while the

Plp1dup mouse has gaps in the blue stain throughout the corpus callosum. Finally, six month old mouse brain sections were observed, and again the same dramatic discontinuous pattern was observed in the Plp1dup brains. Figure 3.11c shows the solid and even myelin pattern in the six month wild-type mice, while the six month

Plp1dup mice show areas of non-myelination within the corpus callosum.

Interestingly, Plp1dup six month mouse sections also showed round holes within the tissue in myelinated areas (figure 3.11c). The holes were seen along the myelin pattern in the corpus callosum. This could be artifact from the histology processing, but if it was then it was probably due to a difference in the Plp1dup mice compared to the wild-type mice at this age because it was only seen in the Plp1dup mice.

Although not as obvious as in the corpus callosum, a disruption in normal myelin formation was also observed in the white matter tracks in the cerebellum.

Figure 3.12 shows the white matter tracks in the cerebellum of wild-type and Plp1dup mice at one month, three months, and six months of age. When observing the myelin pattern in the cerebellum, there were small areas lacking the blue myelin stain in the

Plp1dup mice (figure 3.12) when compared to the wild-type. A few holes similar to

43 those seen within the corpus callosum of the six month Plp1dup mouse sections were also observed in the cerebellum. These holes were also present, but in lower numers, at one and three months. At one month, no difference in the smooth myelin pattern was observed in the Plp1dup mice compared to wild-type mice, but the holes were observed within myelin (figure 3.12a). At three months slightly more difference in the white matter tracks of the cerebellum was observed (figure 3.12b). More holes were seen within the white matter tracks and there were a few areas lacking myelin as shown by small areas of missing blue stain in the Plp1dup mice when compared to the wild-type. Finally, the six month sections, figure 3.12c, shows more differences between wild-type and Plp1dup mice by having more areas of myelin lacking in the cerebellum with the holes throughout the myelin tracks.

To assess whether myelin stability was affected by the duplication in Plp1 locus, spinal cord myelin from six month Plp1dup mice were analyzed by electron microscopy (EM) and compared to six month wild-type mice. Figure 3.13 shows representative data from these analyses and many abnormalities were observed in the

Plp1dup mice sections. Figure 3.13b shows a degenerating axon (Ax with arrow) that was engulfed by a microglial cell (MG), including astrogliosis (AG) in the Plp1dup mice. Figure 3.13c shows a degenerating fiber (DF). Figure 3.13d shows more degenerating fibers (arrows). Figure 3.13e and f show unmyelinated axons (Ax) next to paranode/internode (arrow with PN/IN). These EM images show major abnormalities within the spinal cord myelin in the Plp1dup mouse.

44 EM on the brains of the mice would help acquire a higher resolution and magnification of myelin compaction and stability in the Plp1dup mice. Since brain is a large and complex tissue, it is difficult to find specific regions of interest in the brain with EM because this type of microscopy is at such high magnification. Therefore, correlative microscopy will be used to overcome these difficulties. The Bio-Imaging center at University of Delaware helped to map specific regions of the brain by first using a tile scan technique with confocal microscopy to obtain a full image of the brain section. Still using confocal microscopy, the magnification was increased to identify specific regions of interest and coordinates were obtained by identifying unique areas in the section for EM analysis. This will allow images to be obtained of the corpus callosum and cerebellum at a higher magnification observed in the histological experiments.

Preliminary mapping experiments were completed, and so far the confocal steps of the correlative microscopy were started. Using sagittal sections of the brain, areas of the corpus callosum where axons were seen in cross section were used to assess whether myelin compaction and stability was disrupted in this specific region of the brain. Figure 3.14 shows results of the first mapping experiment with six month wild-type and Plp1dup mice. Using confocal microscopy, tile scans were performed to obtain a full image of the whole section of the brain (figure 3.14a and d) to help map specific regions of interest. Figure 3.14b and e show second tile scan of a specific region selected from the first tile scan at an increased magnification. Finally, figure

45 3.14c and f shows a region from the previous panel with increased magnification.

These images will next be used to define EM coordinates. Figure 3.14 shows the preliminary work of the mapping experiments. Differences between the wild-type and

Plp1dup can be observed in this figure, but the data are preliminary, so conclusions cannot be made from these images. However, this mapping will allow specific regions of the brain to be examined by EM to show detail about the disruption of myelin content in the Plp1dup mouse brain

Table 3.1 Validation of duplication by capillary electrophoresis of multiplex PCR, analysis of gel in figure 3.1b.

Genotyping Analysis 1 2 3 2 4 2 5 2 DNA PLP :DMD neo :DMD region x :DMD puro :DMD Pup 1 1.09 1.21 Pup 2 1.87 0.67 2.04 1.02 Pup 3 0.93 1.02 Pup 4 1.97 0.92 1.84 0.90 Pup 5 1.95 1.12 2.29 1.14 Plp1 dup Male Ctrl 2.00 1.09 2.16 1.08 Plp1 dup Male Ctrl 1.90 0.91 1.87 0.92 Wild type Male Ctrl 0.96 0.89 Wild type Male Ctrl 1.04 1.11 Blue highlighting: Plp1dup mouse 1Region of the Plp1 amplified using the mPLP-F6 and mPLP-R1 primer pair. 2Region of the Dmd amplified using the mDMD28-F and mDMD28-R primer pair. 3Region of the neo gene amplified using the neo-F and neo-R primer pair. 4Region on the proximal end of the duplication (region x) amplified using the 197O15- F and 197O15-R primer pair. 5Region of the puro gene amplified using the MHPP-F14 and MHPP-R5 primer pair. Sequences of these primer pairs are found in table 2.1.

Figure 3.1 Validation of duplication by semi-quantitative multiplex PCR and by amplification across hprt gene a. Diagram of the duplication that was generated in the Plp1dup mouse. Horizontal arrows: positions of the PCR primers, //: large space between neo and DMD gene, thick red line: duplicated portion proximal of Plp1, thick green line: duplicated portion distal of Plp1 b. Sample genotype analysis by multiplex PCR amplification of spleen DNA from a litter of five male mice. Separation of products by agarose gel electrophoresis was used to verify amplification before subjecting samples to capillary electrophoresis. The quantitative data for this gel was obtained by capillary electrophoresis and are shown in table 3.1. Lane M: 100bp marker, lanes 1-5: male pups from a litter, lanes 6 and 7: male controls with the duplication, lanes 8 and 9: wild-type male controls. c. Sample genotype analysis by amplification across the hprt gene. PCR products were separated by agarose gel electrophoresis. Results indicate that samples in lanes 3 and 4 are from Plp1dup male mice, while samples in 1,2, and 5 are wild-type. Lane M: 1kb marker, lanes 1-5: samples from a litter of five male pups, lane 6: normal human DNA control. Grey arrows: represent mDMD28-F and mDMD28-R primer pair, purple arrows: represent the mPLP-F6 and mPLP-R1 primer pair, blue arrows: represent the neo-F and neo-R primer pair, red arrows: represent the 197O15-F and 197O15-R primer pair, green arrows: represent the MHPP-F14 and MHPP-R5 primer pair, black arrows: represent the hHPRT-F2 and hHPRT-R primer pair. All primer pair sequences found in table 2.1.

***

Figure 3.2 Weight difference in Plp1dup males compared to wild-type males at six months of age. Results of the quantitative data for Plp1dup (n=30) show that they weigh an average of 35.2 grams and wild-type (n=30) males weigh an average of 41.2 grams. Plp1dup mice weigh approximately 6 grams less than wild-type males (***, P < 0.001).

Figure 3.3 X-ray analysis of the Plp1dup five month old male mice. The curvature of the spine felt along the spine of Plp1dup mice was analyzed by X-ray analysis of Plp1dup and wild-type males in these three animals. Red arrow points out the curvature of the spine that is observed in one Plp1dup mouse and has not been felt in wild-type mice. The curvature was not felt in the first Plp1dup mouse in this figure and it was not seen in the X-ray.

Figure 3.4 Developmental profile of Plp1+Dm20 mRNA expression in Plp1dup and wild-type mouse brain. The quantitative data shown over the developmental time set for Plp1+Dm20 mRNA expression relative to P1 wild-type. Data were measured by real time qRT-PCR of total RNA isolated from Plp1dup (n=3) and wild- type (n=3) mouse brain at each time point. Results show a triphasic expression pattern in the wild-type animals, with a different expression pattern seen in the Plp1dup mice. These data show an overexpression of Plp1+Dm20 from P12 to P120 with a regulation to normal levels at P21 in the Plp1dup mice.

Figure 3.5 Altered expression of Plp1+Dm20 mRNA and PLP1+DM20 protein in Plp1dup relative to wild-type mice. a. Quantitative data for Plp1+Dm20 mRNA expression in brain measured by real time qRT-PCR of Plp1dup (n=3) and wild-type (n=3) males at each time point indicated. Results were expressed as a percentage of the average of three wild-type mice. P1, P7, P10, and P21 show no difference from wild-type, but significant increases were observed in Plp1dup mice at P12, P14, P60, and P120. b. Quantitative data for PLP1+DM20 protein measured from the Western blot. Results for the Plp1dup mice (n=3) were expressed as a percentage of wild-type (n=3) protein expression. Elevated PLP1+DM20 levels were observed in the Plp1dup at P12, but wild-type levels were observed at older ages. A representative Western blot is shown above the graph. Protein homogenates were prepared from brain at postnatal days indicated. The blot was probed with antibody that recognizes PLP1 and DM20 and an antibody to GAPDH as a loading accuracy control (*p < 0.05, **p < 0.01).

Figure 3.6 Plp1 and Dm20 alternatively spliced transcript levels in the Plp1dup are similar to wild-type. Quantitative data for the ratio of the Plp1 isoform over Plp1+Dm20 mRNA expression in the Plp1dup compared to wild-type levels. The levels of alternative spliced products Plp1 and Dm20 were similar in the Plp1dup mice compared to wild-type controls.

Figure 3.7 Altered mRNA expression of the genes within the duplication, other than Plp1, in Plp1dup mice relative to wild-type mice. a. Diagram of the duplication showing the location of the other genes found within the duplication relative to Plp1. Quantitative data for b. Tceal3 c. Tceal1 d. Morf4l2 e. BC065397 and f. Glra4 mRNA, measured by real time qRT-PCR of mRNA from brain at the time points indicated from Plp1dup (n=3) and wild-type (n=3) male mice. Results are expressed as a percentage of wild-type mice, and elevated levels of mRNA for all genes was observed in Plp1dup mice compared to wild-type mice except for Glra4 (*p < 0.05, **p < 0.01, and ***p < 0.001).

Figure 3.8 Close to wild-type expression levels of CNPase mRNA over the developmental set with lower protein levels in adult Plp1dup mice compared to wild-type mice. a. Quantitative data for CNPase mRNA, measured by real time qRT- PCR of RNA from brain at the time points indicated for Plp1dup (n=3) and wild-type (n=3) male mice. Results are expressed as a percentage of wild-type mice. There was only a difference observed in the Plp1dup mice compared to wild-type at P21 when a slightly lower amount of CNPase expression in the Plp1dup mice was observed. b. Quantitative data for CNPase protein measured from the Western blot. Results for the Plp1dup mice (n=3) are expressed as a percentage of wild-type (n=3). Lower CNPase protein levels were observed in the Plp1dup mice at P60 and P120. A representative Western blot is shown above the graph. Protein homogenates were prepared from brain at postnatal days indicated. The blot was probed with antibody to CNPase and an antibody to GAPDH as a loading accuracy control (*p < 0.05 ** < 0.01).

55 b.

Figure 3.9 Wild-type levels of MAG mRNA expression with altered protein levels in the Plp1dup mice compared to wild-type mice. a. Quantitative data for Mag mRNA, measured by real time qRT-PCR of RNA from brain at the ages indicated of Plp1dup (n=3) and wild-type (n=3) male mice. Results are expressed as a percentage of wild-type mice. There is no significant difference observed in the Plp1dup mice compared to wild-type over the developmental set. b. Quantitative data for MAG protein measured from the Western blot. Results for the Plp1dup mice (n=3) are expressed as a percentage of wild-type (n=3). Lower MAG protein levels were observed in the Plp1dup at P7, P10 and P21. A representative Western blot is shown above the graph. Protein homogenates were prepared from brain at postnatal days indicated. The blot was probed with antibody to MAG and an antibody to GAPDH as a loading accuracy control (*p < 0.05).

56 Figure 3.10 Lower levels of MBP mRNA and protein expression in the Plp1dup mice compared to wild-type mice. a. Quantitative data for Mbp mRNA, measured by real time qRT-PCR of RNA from brain at the ages indicated of Plp1dup (n=3) and wild-type (n=3) male mice. Results are expressed as a percentage of wild-type mice. There were slightly lower levels observed in the Mbp mRNA expression in the Plp1dup mice compared to wild-type at P7, P21, and P60. b. Quantitative data for MBP protein measured from the Western blot. Results for the Plp1dup mice (n=3) are expressed as a percentage of wild-type (n=3). MBP protein begins to be expressed at P20, and lower MBP levels are observed in the Plp1dup at P60, and P120. A representative Western blot is shown above the graph. Protein homogenates were prepared from brain at postnatal days indicated. The blot was probed with antibody to MBP and an antibody to GAPDH as a loading accuracy control (*p < 0.05).

Figure 3.11 Discontinuous myelination pattern in the corpus callosum of Plp1dup mice. Representative image of 5 µm thick brain sections of a. one month, b. three month, and c. six month wild-type and Plp1dup mice stained with Luxol-Fast Blue and PAS. Results show a discontinuous pattern of myelin (blue stain) in the corpus callosum in the Plp1dup mouse compared to the wild-type pattern (n = 3). Holes were observed along myelination pattern in the six month Plp1dup mice. Scale bar: 50 microns.

Figure 3.12 Discontinuous myelination pattern in the white matter tracks of the cerebellum of Plp1dup. Representative image of 5 µm thick brain sections of a. one month, b. three month, and c. six month wild-type and Plp1dup mice stained with Luxol-Fast Blue and PAS. Results show a slight discontinuous pattern of myelin (blue stain) in the cerebellum white matter tracks in the Plp1dup (n = 3) compared to the wild-type (n = 3) myelin pattern. In Plp1dup mouse brain holes were observed along myelination pattern also. Scale bar: 50 microns.

Figure 3.13 Disruption of normal myelin formation in Plp1dup spinal cord. Representative electron micrographs of spinal cord of six month mice a. Wild-type mouse b. Plp1dup mouse with a degenerating axon (Ax w/ arrow) engulfed by a microglial cell (MG) showing astrogliosis (AG) c. Plp1dup mouse showing degenerating fibers (DF) d. Plp1dup mouse showing degenerating fibers (arrows) e. Plp1dup showing an unmyelinated axon (Ax) next to paranode/internode (arrow w/ PN/IN) f. Plp1dup mouse showing an unmyelinated axon (Ax) next to paranode/internode (arrow w/ PN/IN). Scale bar: 1 micron.

Figure 3.14 Preliminary data indicating how specific regions of the brain will be mapped for electron microscopy using confocol microscopy. Images of brain sections of a. - c. six month wild-type and d. - f. six month Plp1dup mice stained with Brain Stain. a. and d. Show the tile scanning technique using a 10x/0.45 objective lens. Images of 15 x 15 were tiled to form a full image. b. and e. Show zoomed into the red square at corpus callosum taken with 10x/0.45 objective at 1-fold zoom and captured 5 x 5 images then tiled to form a full image. c. and f. Show a 2-fold zoom into the blue square of an area of interest showing cross sections of myelin (green) using 10x/0.45 objective. Specific regions of brain will be located and targeted for TEM using this method. Green: fluorescent myelin stain, Red: fluorescent Nissl stain (cell bodies), Blue: DAPI dihydrochloride (DNA).

61 Chapter 4

DISCUSSION

Duplication of the Plp1 locus that includes the entire Plp1 gene is the most frequent cause of PMD. Existing transgenic rodent models containing extra copies of

Plp1 have been characterized, but none of these exactly replicated what is most commonly found among PMD patients. This is the first report of a model that is genetically similar to PMD patients by containing a 217 kb tandem genomic duplication of the X chromosome that not only contains Plp1, but adjacent DNA.

Thus, the duplication also includes regulatory regions as well as five other neighboring genes. Our new model, the Plp1dup mouse, is a more accurate representation of the human duplication at the Plp1 locus that causes PMD compared to previously used models. Conclusions from the characterization of this new mouse model were that the duplication in the Plp1 locus altered the expression of the genes within the duplication, including Plp1. Change in Plp1 expression leads to altered expression of major myelin proteins and disruption of normal myelin formation, which is similar to PMD patients.

The experiments presented in this thesis were performed on a

C57BL/6J and 129S1/SvImJ mixed background mouse line. This mixed background line leads to variation from mouse to mouse due to genetic differences from the two

62 different backgrounds. The mixed background mice makes the data reported here more like what may be observed in a human population, which has a much higher genetic variability. However, the mixed background could account for some of the large error bars and variation from mouse to mouse that was seen in the X-ray analysis

(figure 3.3) and protein expression of MAG (figure 3.9b). Increasing sample size could eliminate some of the error observed in the mixed background line. Our lab is currently working on reproducing this work on a congenic line of C57BL/6, thus allowing for the only difference between the Plp1dup and wild-type mice to be the duplication in the Plp1 locus.

The duplication leads to a distinct abnormal phenotype. An abnormal phenotype was observed in mice during simple weaning and other handling activities.

For instance, Plp1dup mice display problems lifting up their body when being held by their tail, as compared to wild-type litter mates. The data suggest that the duplication causes a development of a curvature in the spine (figure 3.3) and causes noticeable weight difference (figure 3.2). Both slow growth and skeletal deformation are symptoms that have been reported in PMD patients. To determine whether the observed curvature of the spine is due to neuromuscular or skeletal defects, all of the soft tissues must be removed to visualize the spine without influence from the muscle.

Later in life, the Plp1dup mice develop seizures more often than wild-type mice at approximately one year of age. However, only seven Plp1dup mice and seven wild-

63 type littermates were kept until they were a year old. Our lab is currently working to obtain quantitative data on the seizure phenotype.

Previously, it was presumed that Plp1 duplications cause disease as a result of an overexpression of PLP1, but this has not yet been confirmed in patients with PMD (Anderson et al., 1998; I. Griffiths et al., 1998; Karim, Barrie, McCulloch,

Montague, Edgar, Kirkham, Anderson, Nave, Griffiths, & McLaughlin, 2007).

Characterization of the Plp1dup mouse confirms this might be true, but with a much greater complexity than originally thought. The duplication in the Plp1 locus results in an overexpression of Plp1 transcripts in the brain, but not until P12 and older.

Interestingly, at the peak of myelination, P21, there must be a transcriptional or post- transcriptional regulation that keeps specific mRNA levels at P21, so even with a duplication in the Plp1 locus the levels are the same. Results from this report confirmed what was previously shown (Kanfer et al., 1989), that Plp1 displays a triphasic mRNA expression pattern in wild-type mice; a gradual increase in the level of Plp1 transcript leading to a peak at P21, and this is followed by a decline to a plateau level. It is thought that a neuronal signals are responsible for the increased level of myelin related mRNA such as Plp1 (Kanfer et al., 1989). An important regulation at the peak of myelination may cause mRNA levels to be tightly controlled, thus regulating Plp1 mRNA levels to a set amount, even with an increased gene dosage, as observed at P21 in figure 3.4 and 3.5a. Understanding how Plp1 mRNA is regulated at this stage is important mechanism that needs to be elucidated.

64 Terminal complement complexes play a role in the inflammatory pathway and could play an important role in regulating important myelin proteins at a post- transcriptional level if there is activation of the complement system‟s terminal pathway. This could be one possible explanation for the mRNA levels observed in the

Plp1dup mice. The effect of terminal complement complexes in oligodendrocytes on myelin protein gene expression was published, although terminal complement complexes are best known insert into a target as membrane attack complex (MAC) causing cytolysis (Dobrina et al., 2002). Recent evidence has proved that terminal complement complexes in sublytic amounts can have noncytolytic effects on several target cells. Results have specifically shown that sublytic activation of serum C reduced accumulation of mRNA encoding Plp1 and Mbp (Shirazi, Rus, Macklin, &

Shin, 1993). If there is an inflammatory process in the Plp1dup mice due to an overexpression of Plp1, then Plp1 mRNA could be post-transcriptionally regulated explaining that even with an extra gene dosage of Plp1 mRNA levels can be down- regulated to normal levels, as seen at P21 (figure 3.4 and 3.5a). The microglial cells that were shown engulfing the degenerating axons in the spinal cord (figure 3.13b) supports the hypothesis that an inflammatory response could be occurring in Plp1dup mice since these macrophages are the first immune defense in the CNS.

Even though the Plp1 mRNA levels were altered in the brain of mice with a duplication in the Plp1 locus, no change was observed in the Plp1/Dm20 alternative splicing ratio (figure 3.6). This data does not agree with what was previously reported,

65 that a gene duplication causes alteration of the PLP1 to DM20 splicing balance in cultured skin fibroblasts from patients with PMD (Regis et al., 2009). The findings in the Plp1dup mice are not unusual because Plp1 expression was analyzed in brain where it is highly expressed. Only low levels of PLP1+DM20 are expressed in fibroblasts and the ratios obtained from this tissue may be misleading due to the small volume.

PLP1+DM20 protein was overexpressed when the protein was first detected at P12, but after that, the same PLP1+DM20 protein expression levels were seen whether the duplication was present or not. In the #66 hemizygous mouse with seven extra copies of Plp1, it was reported that in spinal cord there was an increase in protein levels at P3 and normal levels at P20 (Karim, Barrie, McCulloch, Montague,

Edgar, Kirkham, Anderson, Nave, Griffiths, & McLaughlin, 2007). These results are similar to the pattern seen in the Plp1dup mouse, where at early protein expression,

P12, there is an overexpression of PLP1+DM20 protein and normal levels at P14, P21,

P60, and P120 (figure 3.5b). The level of PLP1 that is released and inserted into the myelin is a controlled process, and this control could be attributed to translational regulation, rapid turnover, or degradation which explains wild-type protein levels when a duplication exists in the Plp1 locus (Karim et al., 2010).

Results in this report support the idea previously published that PLP1 is highly translationally regulated. It was previously published that the rate of PLP1 translation is decreased in homozygous (14 copies of Plp1) #66 line (Karim et al.,

66 2010). The idea that miRNA could influence PLP1 translation was suggested. This miRNA regulation has shown to be important in the translation of PMP22, which is a peripheral myelin protein that is overexpressed in Charcot-Marie-Tooth disease type

1A (Karim et al., 2010; Verrier et al., 2009). The mouse model characterized in this thesis only provides evidence that supports the hypothesis of translational control since the levels of Plp1 mRNA are shown to be increased due to the genomic duplication, but normal protein levels are observed except at P12, the earliest time of detection.

This report also supports the hypothesis that PLP1 is regulated at the post- translational level. In wild-type oligodendrocytes, approximately 30% of newly synthesized PLP1 is degraded by proteasomes (Karim et al., 2010). This is the same process that has shown to be important in degrading the misfolded PLP1 (McLaughlin et al., 2006). Twice as much PLP1 (approximately 60%) was subject to degradation by this process in oligodendrocytes derived from homozygous (14 copies of Plp1) #66 mouse line compared to the wild-type when degradation was measured by a pulse- chase assay (Karim et al., 2010). The structure of PLP1 in the #66 homozygous mice is thought to be identical to wild-type protein, and therefore does not activate the misfolded protein response. The amount of PLP1 protein inserted into myelin in #66 homozygous mice is about a third of that incorporated in wild-type, so the extra PLP1 is thought to be degraded (Karim et al., 2010). This hypothesis of PLP1 protein regulation through proteosomal degradation could explain the results in the current report, that at an early developmental time point during myelination PLP1 protein was

67 increased compared to wild-type due to increased gene dosage, and this excess PLP1 leads to the later degradation of all the excess protein by proteosomes. Thus, degradation leads to normal PLP1 protein levels at P14, P21, P60, and P120 even with a duplication in the Plp1 locus.

The translational and post-translational regulation hypothesis should be further studied in the Plp1dup mouse. As seen in table 1.1, the previous mouse models were only analyzed at a single time point and showed that PLP1 is either overexpressed or underexpressed usually dependent upon whether they are hemizygous or homozygous. In the Plp1dup mouse model, the duplication is similar to what patients will have, and the developmental expression pattern of PLP1 during myelination was observed in this report showing its complex expression (figures 3.4 and 3.5), and a greater understanding of PLP1 regulation could be elucidated using this model. It is possible that the duplication alters the regulation and insertion of PLP1 into the myelin. In order to determine this, a localization assays must be performed in the Plp1dup mouse to visualize if PLP1 is inserted into myelin or is elsewhere in the cell. Since results in this study show that PLP1 protein levels are regulated to wild- type levels during myelination and into adulthood, it is important to elucidate the location of this protein to understand the improper formation of myelin seen in figures

3.11, 3.12, and 3.13. There is a regulation allowing for the proper amount of PLP1 protein in whole brain, but the issue might exist in the regulation of the PLP1 protein being properly inserted into the myelin.

68 Timing of the myelin development could be crucial and is yet another explanation of the improper myelin formation observed in the Plp1dup mice. In wild- type mice, the peak of myelin accumulation is at P21, and this is also where Plp1 expression reaches a peak (Kanfer et al., 1989). However, in the Plp1dup mice Plp1 expression reaches a peak at an earlier age, P14 (figure 3.4), and this could be detrimental to proper myelin formation. The lab will have to investigate myelin formation by measuring the accumulation of myelin, then it could be identified if myelin is reaching a peak accumulation at an earlier time in the Plp1dup mice. If this timing of myelin formation is occurring sooner, this could be cause the improper formation that was observed in the Plp1dup and cause the degeneration and hypomyelination seen in these animals.

Improper myelin formation due to the duplication in the Plp1 locus could account for the decrease in other myelin proteins: MBP, MAG, and CNPase (Figures

3.8, 3.9, and 3.10). Normally, PLP1 associates with cholesterol and other lipids to form myelin rafts during trafficking through the Golgi complex (Simons et al., 2000).

Excessive amounts of PLP1 proteins have been shown previously to disrupt this pathway, and excessive amounts of normal PLP1 protein accumulate in late endosome and lysosomal compartments of rodent cells (Simons et al., 2002). Therefore, overexpression of PLP1 causes an improper transport and assembly of myelin constituents, and would explain the decrease in levels of other myelin proteins and the improper formation of myelin. The holes observed in the pathology experiments could

69 be where lipids and protein, including excess PLP1, are not being effectively inserted into myelin and are being washed away during processing of the slides. If this hypothesis is true, it would further explain why protein levels of PLP1 are observed to be normal with the duplication, but improper myelin formation is observed.

Interestingly, MBP has slightly lower levels of mRNA expression in the

Plp1dup mice compared to wild-type mice and a decrease in protein expression due to the duplication. MBP was shown to be the only myelin protein to be essential for formation of CNS myelin (Boggs, 2006). Interactions have been shown to exist between MBP and PLP1 by using a microtitre well binding assay and affinity chromatography, and these studies showed that PLP1 binds to MBP, but not to any other CNS myelin proteins (Edwards, Ross, Ulmer, & Braun, 1989; Wood, Vella, &

Moscarello, 1984). It has been shown that MBP influences the partitioning of PLP1 in lipid rafts (Fitzner et al., 2006). When PLP1 is ready for raft formation, a signal transduction cascade could be sent to express MBP to help with the partitioning of these lipid rafts. However, when PLP1 is overexpressed, the excess proteins along with other myelin constituents that normally associate with PLP1 in these rafts are found in the lysomal and endosomal vesicles. If this raft formation is important in signaling for MBP, then it would explain why an increased gene dosage of Plp1 would cause decreased levels of MBP because it would get degraded in the lysosome and endosomes.

70 The mRNA expression of MAG and CNPase does not correlate with their protein expression levels in the Plp1dup mouse. MAG and CNPase have generally normal mRNA expression levels during myelination into adulthood in the Plp1dup mice (figure 3.8 and 3.9). However, CNPase has a significant decrease of protein in the adult Plp1dup mouse brain compared to wild-type mice as mice age (figure 3.8b).

MAG does not have a simple decrease in expression as mice age like CNPase does, but at P7, P10, and P21 MAG protein is decreased due to an extra dose of Plp1 (figure

3.9b). This lack of correlation between mRNA and protein expression in MAG and

CNPase could be explained by the importance of MBP in the formation of myelin.

This report shows that the Plp1dup mouse has low mRNA and protein levels of MBP, which could influence the myelin that is formed in these animals since MBP is necessary for myelin formation (Boggs, 2006). This hypothesis could be tested by culturing oligodendrocytes from Plp1dup mice and transfect them with an overexpressing vector with MBP to increase the expression of this gene in the Plp1dup animals to see if MAG and CNPase levels return to normal.

Since Plp1 was shown to be sensitive to gene dosage in such a complex way, other genes within the duplicated region might also play a critical role in the phenotype of PMD. The mRNA expression of the five genes other than Plp1 that are in the Plp1dup duplication was analyzed and four of the five were found to be increased about two-fold. Tceal1, Tceal3, Morf4l2, and Glra4 are commonly found within the duplicated region of patients, so it is important to understand if the function

71 of these genes is affected by overexpression. Tceal1, Tceal3, and Morf4l2 are expressed from 1.5 to 3.5 fold more than wild-type when included in the chromosomal duplication in the Plp1 locus (figure 3.7). Interestingly, none of the overexpressed genes show normal levels at P21, as seen in Plp1 (figure 3.5a). This result leads to the hypothesis that there must be a regulation that is specific to Plp1 transcription at this time point. Although altered expression of Tceal1, Tceal3, and Morf4l2 has been reported to be involved with tumors it has not been reported that PMD patients have an increase incidence of tumors. The Plp1dup mice also do not develop tumors and we did not detect a phenotype that would be attributed to overexpression of Tceal1,

Tceal3, and Morf4l2. However, the fact that an extra dose of these genes could contribute in some way to phenotype in these patients cannot yet be eliminated.

BC065397 is a hypothetical gene that is thought to be part of Ras-like GTPase superfamily, and found at Xq22 in the mouse genome. This hypothetical gene is overexpressed with an extra gene dosage. This hypothetical gene has not yet been reported to be in the human genome, but cannot be ignored when characterizing this new mouse model. Unlike the other duplicated genes, Glra4 was not shown to be sensitive to gene dosage (figure 3.7f). In fact, it was previously reported that this gene is not considered to be involved in the phenotype of PMD (Muncke et al., 2004).

Actually, unlike other glycine receptor subunit genes, Glra4 harbors a premature stop codon within exon 9, so the forth transmembrane domain is not translated. Thus,

Glra4 is thought to be a non-functional pseudogene (Muncke et al., 2004).

72 In conclusion, this study has elaborated on the simple idea that an increased gene dosage of the Plp1 gene leads to an overexpression of mRNA and protein levels. Interestingly, PLP1 has a complex regulation, leading to not just a simple overexpression due to a genomic duplication, but a complicated altered PLP1 expression throughout the development into adulthood. This model will allow for the study of how, not only Plp1, but the neighboring DNA, including the other genes, is important to understanding how this type of genomic duplication leads to PMD. Other genes that have been shown to be overexpressed due to an extra gene dosage exist within genomic duplications found among PMD patients and should be taken into account when trying to understand how a duplication in the Plp1 locus leads to PMD.

The Plp1dup mouse model is a more accurate representation of PMD compared to previously reported models and the initial characterizations presented here are the early steps of elucidating the role of a genomic duplication in the Plp1 locus on the

PMD phenotype.

73 Appendix A

NORMALIZATION OF REAL-TIME RT-PCR

Due to the sensitivity in real-time PCR results, finding the most stably expressed control gene in mouse brain tissue along the developmental time course studied in this thesis was very important. Increasing evidence has been shown that normalizing to a single reference gene introduces large and variable errors into analysis. The geNorm system was used for selecting the best candidate reference genes for the present data in this thesis. The underlying principles and formulas for this system have been described (Vandesompele et al., 2002). The table below shows the 12 reference genes measured with our samples (PrimerDesign).

Mouse geNorm reference gene kit 1 Mus musculus actin, beta, cytoplasmic (ACTB), mRNA. 2 Mus musculus glyceraldehyde-3-phosphate dehydrogenase (GAPDH), mRNA. 3 Mus musculus ubiquitin C (UBC), mRNA. 4 Mus musculus beta-2 microglobulin (B2M), mRNA. 5 Mus musculus phospholipase A2 (YWHAZ), mRNA. 6 Mus musculus ribosomal protein L13a (RPL13A), mRNA. 7 Mus musculus calnexin (CANX), mRNA. 8 Mus musculus cytochrome c-1 (CYC1), mRNA.

9 Mus musculus succinate dehydrogenase complex, subunit A (SDHA), mRNA. 10 Mus musculus 18S rRNA gene

11 Mus musculus eukaryotic translation initiation factor 4A2 (EIF4A2), mRNA. 12 Mus musculus ATP synthase subunit (ATP5B), mRNA.

Real-time PCR was performed following the PrimerDesign geNorm

SYBRgreen protocol (PrimerDesign Ltd., Southhampton, Hampshire UK). Ct values

74 were converted into relative quantification data using the deltaCT method and prepared for input into the geNorm for Windows VBA applet for Microsoft Excel

200/XP/2003 (version3.5), available for download at http://medgen.ugent.be/~jvdesomp/genorm/. The manufacturer‟s protocol was followed to analyze data using the software.

The first graph that was generated entitled average expression stability values of remaining control genes indicated the average expression stability value M of reference genes, starting from the least stable gene at the left, increasing expression stability ending with the two most stable genes on the right. Thus, the conclusion was reached that CANX and CYC1 are the two most stable genes among our developmental mouse brain RNA samples. The second graph generated entitled determination of the optimal number of control genes for normalization illustrated levels of variation in average reference gene stability with the addition of each reference gene to the equation. The graph shows the two most stably expressed genes on the left, including a third, fourth, and fifth gene etc. at each data point moving to the right. This is the pairwise variation V. PrimerDesign protocol (PrimerDesign Ltd.) recommends a V score below 0.15 as an ideal normalization. Among our samples,

0.077 is the V score using two normalizing genes, which will give ideal data.

Therefore, these analyses lead to the selection of CANX and CYC1 for the normalization of RNA from mouse brain along the developmental set in real-time analysis.

76 Appendix B

INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE APPROVAL

Institutional

Animal Care and Use Committee

MEMORANDUM

DATE: September 9, 2010

TO: Grace M. Hobson, Ph.D.

FROM: Paul T. Fawcett, Ph.D.

SUBJECT: “Mouse model of PMD due to duplication of PLP1” NBR-2007- 007

The Institutional Animal Care and Use Committee (IACUC) have reviewed the above referenced protocol for approval, and the following decision has been made:

Action: Renewal approved

Date of Action: September 9, 2010

Approval Period: September 9, 2010 – September 8, 2011

Protocol Approval Number: NBR-2007-007

77 Approved Number of Animals: Mus – mixed background (C57BL/6J and 129sv/ev) Yearly No. -199; Project Total -596 Mus – C57BL/6J Yearly No. - 233; Project Total -700

Please submit your Biosafety Classification form electronically to the Alfred I. duPont Hospital for Children Institutional Biosafety Committee via the link: http://www.nemours.org/research/committee/ibc.html Please note that the study cannot begin until the Office of Regulatory Compliance in Research Administration has received all approvals.

Please maintain this approval with your project records. A tally of the number of animals approved and the number ordered for the project will be maintained in the Life Science Center. If changes occur in your protocol or if you require more animals than approved, and amendment to your protocol will need to be submitted for consideration

If you have any questions regarding this memorandum, please contact Paul T. Fawcett, Ph.D. at x 6776 or email: [email protected].

78 References

Adams, D. J., Biggs, P. J., Cox, T., Davies, R., van der Weyden, L., Jonkers, J., Bradley, A. (2004). Mutagenic insertion and chromosome engineering resource (MICER). Nature Genetics, 36(8), 867-871.

Anderson, T. J., Klugmann, M., Thomson, C. E., Schneider, A., Readhead, C., Nave, K. A., & Griffiths, I. R. (1999). Distinct phenotypes associated with increasing dosage of the PLP gene: Implications for CMT1A due to PMP22 gene duplication. Annals of the New York Academy of Sciences, 883, 234-246.

Anderson, T. J., Schneider, A., Barrie, J. A., Klugmann, M., McCulloch, M. C., Kirkham, D., Griffiths, I. R. (1998). Late-onset neurodegeneration in mice with increased dosage of the proteolipid protein gene. The Journal of Comparative Neurology, 394(4), 506-519.

Arroyo, E. J., Xu, T., Poliak, S., Watson, M., Peles, E., & Scherer, S. S. (2001). Internodal specializations of myelinated axons in the central nervous system. Cell and Tissue Research, 305(1), 53-66.

Bauer, J., Bradl, M., Klein, M., Leisser, M., Deckwerth, T. L., Wekerle, H., & Lassmann, H. (2002). Endoplasmic reticulum stress in PLP-overexpressing transgenic rats: Gray matter oligodendrocytes are more vulnerable than white matter oligodendrocytes. Journal of Neuropathology and Experimental Neurology, 61(1), 12-22.

Baumann, N., & Pham-Dinh, D. (2001). Biology of oligodendrocyte and myelin in the mammalian central nervous system. Physiological Reviews, 81(2), 871-927.

Boggs, J. M. (2006). Myelin basic protein: A multifunctional protein. Cellular and Molecular Life Sciences, 63(17), 1945-1961.

79 Bradl, M., Bauer, J., Inomata, T., Zielasek, J., Nave, K. A., Toyka, K., Wekerle, H. (1999). Transgenic lewis rats overexpressing the proteolipid protein gene: Myelin degeneration and its effect on T cell-mediated experimental autoimmune encephalomyelitis. Acta Neuropathologica, 97(6), 595-606.

Brady, S. T., Witt, A. S., Kirkpatrick, L. L., de Waegh, S. M., Readhead, C., Tu, P. H., & Lee, V. M. (1999). Formation of compact myelin is required for maturation of the axonal cytoskeleton. The Journal of Neuroscience, 19(17), 7278-7288.

Caro, P. A., & Marks, H. G. (1990). Magnetic resonance imaging and computed tomography in Pelizaeus-Merzbacher Disease. Magnetic Resonance Imaging, 8(6), 791-796.

Cheon, J. E., Kim, I. O., Hwang, Y. S., Kim, K. J., Wang, K. C., Cho, B. K., Yeon, K. M. (2002). Leukodystrophy in children: A pictorial review of MR imaging features. Radiographics, 22(3), 461-476.

Costantino-Ceccarini, E., & Morell, P. (1972). Biosynthesis of brain sphingolipids and myelin accumulation in the mouse. Lipids, 7(10), 656-659.

Dobrina, A., Pausa, M., Fischetti, F., Bulla, R., Vecile, E., Ferrero, E., Tedesco, F. (2002). Cytolytically inactive terminal complement complex causes transendothelial migration of polymorphonuclear leukocytes in vitro and in vivo. Blood, 99(1), 185-192.

Dubois-Dalcq, M., Behar, T., Hudson, L., & Lazzarini, R. A. (1986). Emergence of three myelin proteins in oligodendrocytes cultured without neurons. The Journal of Cell Biology, 102(2), 384-392.

Edwards, A. M., Ross, N. W., Ulmer, J. B., & Braun, P. E. (1989). Interaction of myelin basic protein and proteolipid protein. Journal of Neuroscience Research, 22(1), 97-102.

Ellis, D., & Malcolm, S. (1994). Proteolipid protein gene dosage effect in Pelizaeus- Merzbacher Disease. Nature Genetics, 6(4), 333-334.

80 Fitzner, D., Schneider, A., Kippert, A., Mobius, W., Willig, K. I., Hell, S. W., Simons, M. (2006). Myelin basic protein-dependent plasma membrane reorganization in the formation of myelin. The EMBO Journal, 25(21), 5037-5048.

Fujimoto, K., Roots, B. I., Burton, R. M., & Agrawal, H. C. (1976). Morphological and biochemical characterization of light and heavy myelin isolated from developing rat brain. Biochimica Et Biophysica Acta, 426(4), 659-668.

Garbern, J. Y. (2007). Pelizaeus-Merzbacher Disease: Genetic and cellular pathogenesis. Cellular and Molecular Life Sciences, 64(1), 50-65.

Garbern, J. Y., Yool, D. A., Moore, G. J., Wilds, I. B., Faulk, M. W., Klugmann, M., Griffiths, I. R. (2002). Patients lacking the major CNS myelin protein, proteolipid protein 1, develop length-dependent axonal degeneration in the absence of demyelination and inflammation. Brain, 125(Pt 3), 551-561.

Gencic, S., Abuelo, D., Ambler, M., & Hudson, L. D. (1989). Pelizaeus-Merzbacher Disease: An X-linked neurologic disorder of myelin metabolism with a novel mutation in the gene encoding proteolipid protein. American Journal of Human Genetics, 45(3), 435-442.

Gow, A., & Lazzarini, R. A. (1996). A cellular mechanism governing the severity of Pelizaeus-Merzbacher Disease. Nature Genetics, 13(4), 422-428.

Griffiths, I., Klugmann, M., Anderson, T., Yool, D., Thomson, C., Schwab, M. H., Nave, K. A. (1998). Axonal swellings and degeneration in mice lacking the major proteolipid of myelin. Science, 280(5369), 1610-1613.

Griffiths, I. R., Montague, P., & Dickinson, P. (1995). The proteolipid protein gene. Neuropathology and Applied Neurobiology, 21(2), 85-96.

Hudson, L. D. (2003). Pelizaeus-Merzbacher Disease and spastic paraplegia type 2: Two faces of myelin loss from mutations in the same gene. Journal of Child Neurology, 18(9), 616-624.

81 Hudson, L. D., Puckett, C., Berndt, J., Chan, J., & Gencic, S. (1989). Mutation of the proteolipid protein gene PLP in a human X chromosome-linked myelin disorder. Proceedings of the National Academy of Sciences of the United States of America, 86(20), 8128-8131.

Huxley, A. F., & Stampfli, R. (1949). Evidence for saltatory conduction in peripheral myelinated nerve fibres. The Journal of Physiology, 108(3), 315-339.

Inoue, K., Osaka, H., Imaizumi, K., Nezu, A., Takanashi, J., Arii, J., Lupski, J. R. (1999). Proteolipid protein gene duplications causing Pelizaeus-Merzbacher Disease: Molecular mechanism and phenotypic manifestations. Annals of Neurology, 45(5), 624-632.

Inoue, K., Osaka, H., Sugiyama, N., Kawanishi, C., Onishi, H., Nezu, A., Kosaka, K. (1996). A duplicated PLP gene causing Pelizaeus-Merzbacher Disease detected by comparative multiplex PCR. American Journal of Human Genetics, 59(1), 32-39.

Inoue, Y., Kagawa, T., Matsumura, Y., Ikenaka, K., & Mikoshiba, K. (1996). Cell death of oligodendrocytes or demyelination induced by overexpression of proteolipid protein depending on expressed gene dosage. Neuroscience Research, 25(2), 161-172.

Kagawa, T., Ikenaka, K., Inoue, Y., Kuriyama, S., Tsujii, T., Nakao, J., Mikoshiba, K. (1994). Glial cell degeneration and hypomyelination caused by overexpression of myelin proteolipid protein gene. Neuron, 13(2), 427-442.

Kanfer, J., Parenty, M., Goujet-Zalc, C., Monge, M., Bernier, L., Campagnoni, A. T., Zalc, B. (1989). Developmental expression of myelin proteolipid, basic protein, and 2',3'-cyclic nucleotide 3'-phosphodiesterase transcripts in different rat brain regions. Journal of Molecular Neuroscience, 1(1), 39-46.

Karim, S. A., Barrie, J. A., McCulloch, M. C., Montague, P., Edgar, J. M., Iden, D. L., McLaughlin, M. (2010). PLP1 expression and turnover in a transgenic mouse model of Pelizaeus-Merzbacher Disease. Glia, 58(14), 1727-1738.

82 Karim, S. A., Barrie, J. A., McCulloch, M. C., Montague, P., Edgar, J. M., Kirkham, D., McLaughlin, M. (2007). PLP overexpression perturbs myelin protein composition and myelination in a mouse model of Pelizaeus-Merzbacher Disease. Glia, 55(4), 341-351.

Koeppen, A. H., & Robitaille, Y. (2002). Pelizaeus-Merzbacher Disease. Journal of Neuropathology and Experimental Neurology, 61(9), 747-759.

Koeppen, A. H., Ronca, N. A., Greenfield, E. A., & Hans, M. B. (1987). Defective biosynthesis of proteolipid protein in Pelizaeus-Merzbacher Disease. Annals of Neurology, 21(2), 159-170.

Li, Z., Yu, T., Morishima, M., Pao, A., LaDuca, J., Conroy, J.,Yu, Y. E. (2007). Duplication of the entire 22.9 mb human chromosome 21 syntenic region on mouse chromosome 16 causes cardiovascular and gastrointestinal abnormalities. Human Molecular Genetics, 16(11), 1359-1366.

McLaughlin, M., Barrie, J. A., Karim, S., Montague, P., Edgar, J. M., Kirkham, D., Griffiths, I. R. (2006). Processing of PLP in a model of Pelizaeus-Merzbacher Disease/SPG2 due to the rumpshaker mutation. Glia, 53(7), 715-722.

Merzbacher, L. (1910) Eine eigenarige familiär-hereditare Erkrankungsform (Aplasia axialis extra-corticalis congenita). Z Gesamte Neurol Psychiatr. 3, 1–138.

Meusser, B., Hirsch, C., Jarosch, E., & Sommer, T. (2005). ERAD: The long road to destruction. Nature Cell Biology, 7(8), 766-772.

Monge, M., Kadiiski, D., Jacque, C. M., & Zalc, B. (1986). Oligodendroglial expression and deposition of four major myelin constituents in the myelin sheath during development. an in vivo study. Developmental Neuroscience, 8(4), 222- 235.

Morello, D., Dautigny, A., Pham-Dinh, D., & Jolles, P. (1986). Myelin proteolipid protein (PLP and DM-20) transcripts are deleted in jimpy mutant mice. The EMBO Journal, 5(13), 3489-3493.

83 Muncke, N., Wogatzky, B. S., Breuning, M., Sistermans, E. A., Endris, V., Ross, M., Rappold, G. (2004). Position effect on PLP1 may cause a subset of Pelizaeus- Merzbacher Disease symptoms. Journal of Medical Genetics, 41(12), e121.

Nave, K. A., Bloom, F. E., & Milner, R. J. (1987). A single nucleotide difference in the gene for myelin proteolipid protein defines the jimpy mutation in mouse. Journal of Neurochemistry, 49(6), 1873-1877.

Pelizaeus F. Uber eine eigenthümliche Form Spastischer Lähmung mit Cerebralerschinungen auf hereditärer Grundlage (Multiple Sklerose). Arch Psychiatr Nervenkr.1885;16:698-710

Pfeiffer, S. E., Warrington, A. E., & Bansal, R. (1993). The oligodendrocyte and its many cellular processes. Trends in Cell Biology, 3(6), 191-197.

Popot, J. L., Pham Dinh, D., & Dautigny, A. (1991). Major myelin proteolipid: The 4- alpha-helix topology. The Journal of Membrane Biology, 120(3), 233-246.

Readhead, C., Schneider, A., Griffiths, I., & Nave, K. A. (1994). Premature arrest of myelin formation in transgenic mice with increased proteolipid protein gene dosage. Neuron, 12(3), 583-595.

Regis, S., Grossi, S., Corsolini, F., Biancheri, R., & Filocamo, M. (2009). PLP1 gene duplication causes overexpression and alteration of the PLP1 splicing balance in fibroblasts from Pelizaeus-Merzbacher Disease patients. Biochimica Et Biophysica Acta, 1792(6), 548-554.

Scherer, S. S., Vogelbacker, H. H., & Kamholz, J. (1992). Axons modulate the expression of proteolipid protein in the CNS. Journal of Neuroscience Research, 32(2), 138-148.

Schneider, A., Lander, H., Schulz, G., Wolburg, H., Nave, K. A., Schulz, J. B., & Simons, M. (2005). Palmitoylation is a sorting determinant for transport to the myelin membrane. Journal of Cell Science, 118(Pt 11), 2415-2423.

84 Seitelberger, F. (1954) Die Pelizaeus-Merzbachersche Krankheit. Klinischanatomische Untersuchungen zum Problem ihrer Stellung unter den diffusen Sklerosen. Wien Z Nervenheilkd. 9, 228–28

Sheehan, D.C. and Hrapchak, B.B., 1980. Theory and Practice of Histotechnology. (pp. 46-48). St. Louis, MO: Mosby.

Shirazi, Y., Rus, H. G., Macklin, W. B., & Shin, M. L. (1993). Enhanced degradation of messenger RNA encoding myelin proteins by terminal complement complexes in oligodendrocytes. Journal of Immunology, 150(10), 4581-4590.

Simons, M., Kramer, E. M., Macchi, P., Rathke-Hartlieb, S., Trotter, J., Nave, K. A., & Schulz, J. B. (2002). Overexpression of the myelin proteolipid protein leads to accumulation of cholesterol and proteolipid protein in endosomes/lysosomes: Implications for Pelizaeus-Merzbacher Disease. The Journal of Cell Biology, 157(2), 327-336.

Simons, M., Kramer, E. M., Thiele, C., Stoffel, W., & Trotter, J. (2000). Assembly of myelin by association of proteolipid protein with cholesterol- and galactosylceramide-rich membrane domains. The Journal of Cell Biology, 151(1), 143-154.

Sistermans, E. A., de Coo, R. F., De Wijs, I. J., & Van Oost, B. A. (1998). Duplication of the proteolipid protein gene is the major cause of Pelizaeus-Merzbacher Disease. Neurology, 50(6), 1749-1754.

Timsit, S., Martinez, S., Allinquant, B., Peyron, F., Puelles, L., & Zalc, B. (1995). Oligodendrocytes originate in a restricted zone of the embryonic ventral neural tube defined by DM-20 mRNA expression. The Journal of Neuroscience, 15(2), 1012-1024.

Trofatter, J. A., Dlouhy, S. R., DeMyer, W., Conneally, P. M., & Hodes, M. E. (1989). Pelizaeus-Merzbacher Disease: Tight linkage to proteolipid protein gene exon variant. Proceedings of the National Academy of Sciences of the United States of America, 86(23), 9427-9430.

85 van der Knaap, M. S., & Valk, J. (1989). The reflection of histology in MR imaging of Pelizaeus-Merzbacher Disease. American Journal of Neuroradiology, 10(1), 99- 103.

Vandesompele, J., De Preter, K., Pattyn, F., Poppe, B., Van Roy, N., De Paepe, A., & Speleman, F. (2002). Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biology, 3(7),

Verrier, J. D., Lau, P., Hudson, L., Murashov, A. K., Renne, R., & Notterpek, L. (2009). Peripheral myelin protein 22 is regulated post-transcriptionally by miRNA-29a. Glia, 57(12), 1265-1279.

Walz, K., Paylor, R., Yan, J., Bi, W., & Lupski, J. R. (2006). Rai1 duplication causes physical and behavioral phenotypes in a mouse model of dup(17)(p11.2p11.2). The Journal of Clinical Investigation, 116(11), 3035-3041.

Willard, H. F., & Riordan, J. R. (1985). Assignment of the gene for myelin proteolipid protein to the X chromosome: Implications for X-linked myelin disorders. Science, 230(4728), 940-942.

Wolf, N. I., Sistermans, E. A., Cundall, M., Hobson, G. M., Davis-Williams, A. P., Palmer, R.,Woodward, K. J. (2005). Three or more copies of the proteolipid protein gene PLP1 cause severe Pelizaeus-Merzbacher Disease. Brain, 128(Pt 4), 743-751.

Wood, D. D., Vella, G. J., & Moscarello, M. A. (1984). Interaction between human myelin basic protein and lipophilin. Neurochemical Research, 9(10), 1523-1531.

Woodward, K., Kendall, E., Vetrie, D., & Malcolm, S. (1998). Pelizaeus-Merzbacher Disease: Identification of Xq22 proteolipid-protein duplications and characterization of breakpoints by interphase FISH. American Journal of Human Genetics, 63(1), 207-217.

86 Woodward, K. J., Cundall, M., Sperle, K., Sistermans, E. A., Ross, M., Howell, G., Hobson, G. M. (2005). Heterogeneous duplications in patients with Pelizaeus- Merzbacher Disease suggest a mechanism of coupled homologous and nonhomologous recombination. American Journal of Human Genetics, 77(6), 966-987.

Yu, Y., & Bradley, A. (2001). Engineering chromosomal rearrangements in mice. Nature Reviews Genetics, 2(10), 780-790.